|| EUROPEAN COMMISSION ||

Brussels, 17.7.2012

SWD(2012) 200 final

VOLUME II (Annexes)

 

COMMISSION STAFF WORKING DOCUMENT

Impact assessment report on the revision of the “Clinical Trials Directive” 2001/20/EC

Accompanying the document

Proposal for a Regulation of the European Parliament and of the Council

on clinical trials on medicinal products for human use, and repealing Directive 2001/20/EC

{COM(2012) 369 final} {SWD(2012) 201 final}

              Annex 1: Clinical Trials and the Clinical Trials Directive

(1) Clinical trials

Clinical trials are performed in many different contexts. Traditionally, these are referred to as ‘phases’. While there is no universally agreed terminology, the phases can be defined as follows:[1]

           Phase I (human pharmacology): Initial trials provide an early evaluation of short-term safety and tolerability and can provide pharmacodynamic and pharmacokinetic information needed to choose a suitable dosage range and administration schedule for initial exploratory therapeutic trials.

           Phase II (therapeutic exploratory): Phase II is usually considered to begin with the start of studies in which the primary objective is to explore therapeutic efficiency in patients. In addition, additional information on the safety profile of a compound is gathered.

           Phase III (therapeutic confirmatory): Phase III is usually considered to begin with the start of studies in which the primary objective is to demonstrate or confirm therapeutic benefits. In addition, additional information on the safety profile of a compound is gathered.

           Phase IV (therapeutic use): Phase IV begins after authorisation of the medicinal product. Therapeutic use studies go beyond the prior demonstration of the safety and efficacy of the medicine and dose definition. Phase IV covers all studies (other than routine surveillance) performed after drug approval and related to the approved indication. They are studies that were not considered necessary for approval but are often important for optimising use of the medicinal product. They may be of any type, but should have valid scientific objectives. They commonly include additional drug-drug interaction, dose-response or safety studies and studies to support use for the approved indication, e.g. mortality/morbidity studies.

(2) The Clinical Trials Directive

Prior to the entry into force of the Clinical Trials Directive, the rules for performing clinical trials varied significantly in the Union. Since 2004, clinical trials performed in the EU are regulated by the Clinical Trials Directive. The primary purpose of this Directive is to:

· Protect the health and safety of participants in clinical trials;

· Ensure the reliability and robustness of data generated in clinical trials; and

· Simplify and harmonise the administrative provisions governing clinical trials in order to allow cost-efficient clinical research.[2]

Since the Clinical Trials Directive entered into force, it has been supplemented by a Commission Directive[3] setting out the principles of good clinical practice (GCP). A multitude of other guidance documents have been published in EudraLex, Volume 10,[4] including the Guideline on ‘Good Clinical Practice — ICH E6’. This guideline was agreed under the auspices of the ICH and is, de facto, recognised worldwide as the standard applicable to GCP.

In terms of substance, these Union rules aim at establishing, inter alia:

· Procedures for applications to conduct a clinical trial and authorisation of a clinical trial by the national competent authority (NCA) and Ethics Committee;

· Requirements for a clinical trial, including rules for protection of participants;[5]

· Rules on reporting adverse events, in particular ‘suspected unexpected serious adverse reactions’ (SUSARs), during the clinical trial;

· Rules on the manufacturing, importation and labelling of the ‘investigational medicinal product’ (IMP); and

· Rules on inspection of clinical trial sites.

As a result of this harmonisation, today, clinical data generated anywhere in the EU is accepted, as regards subject rights and safety, as well as data robustness and reliability.

The Clinical Trials Directive does not address the question of whether and how the result of a clinical trial can be used, for example in an application for a marketing authorisation for a medicinal product. Instead, this is regulated in Directive 2001/83/EC of the European Parliament and of the Council of 6 November 2001 on the Community code relating to medicinal products for human use[6] (the ‘Community Code’). The Community Code stipulates that all clinical trials performed in the EU and submitted as part of an application for marketing authorisation must comply with the Clinical Trials Directive. If the clinical trials are performed in non-EU countries, they must comply with rules and principles equivalent to those laid down by the Directive.

The Clinical Trials Directive applies only to ‘interventional trials’, but not to ‘non-interventional’ studies.[7] Non-interventional studies are trials where the medicinal product is prescribed in the usual manner in accordance with the terms of the marketing authorisation, the assignment of the patient to a particular therapeutic strategy is not decided in advance by a trial protocol but falls within current practice, prescription of the medicine is clearly separated from the decision to include the patient in the study, no additional diagnostic or monitoring procedures are applied to the patients and epidemiological methods are used to analyse the data collected.

The reason for excluding non-interventional trials from the scope of the Directive is that they typically pose a lower risk than interventional trials. In addition, this restriction is meant to exclude medical activities which are normal clinical practice and, as such, part of the general medical surveillance of a patient.

The results of observational trials cannot be taken as the basis for an application for marketing authorisation.

The Clinical Trials Directive provides for a database — EudraCT — which contains protocol-related information on clinical trials performed in the EU or contained in a ‘paediatrics investigation plan’ (PIP). The sponsor submits this information on a specially designed form, together with the application for authorisation of a clinical trial, to the NCA of the Member State concerned, which forwards this information to EudraCT. EudraCT is managed by the Agency.

(3) Criticism of the Clinical Trials Directive

The Clinical Trials Directive is the most heavily criticised piece of legislation of the entire EU acquis for pharmaceuticals. The severe criticism is voiced by all stakeholders and political actors ‑ patients, industry, and academic research, Member States, Union institutions ‑ and has been re-iterated and stressed during the various consultations referred to in point Error! Reference source not found.. Examples are:

           Patient organisations: The European Cancer Patient Coalition, in its response to the 2009/10 public consultation stressed that "[The Clinical Trials Directive] has severely hampered cancer research in Europe, and threatens to further destruct existing multi-national research networks which have been established prior to the Clinical Trials Directive. […] The Clinical Trials Directive has created many additional burdens for the conduction of trials, while it did not meet the primary objective of harmonizing and simplifying the legislation in the Member States."[8]

           Industry: The European Federation of pharmaceutical industries and associations (EFPIA) considers that "the European Clinical Trials Directive has added administrative and regulatory constraints in some EU countries where there weren't any such measures or where these were set a lower level without - until now - bringing the tangible benefits of a real harmonisation of the framework conditions to conduct clinical trials across Europe (despite the fact that this initially was the intended goal). In this context, large-scale multi-centred clinical trials are very difficult and cumbersome to operate in Europe, whatever the disease area or medical indication, which may translate into long delays and higher costs."[9] Regarding SMEs, the European Confederation of Pharmaceutical Entrepreneurs (EUCOPE), who has a large membership of SMEs, has stressed during the 2009/10 public consultation that 'Difficulties for SMEs are in parts similar to those of larger companies. However, there is a higher burden for SMEs due to the increased need of staff for preparation and management of clinical trials […]. This leads to an overall increase in resources required for the performance of clinical trials in the new regulatory framework which is especially burdensome for SMEs.[10] This was also highlighted by other respondents, stressing that the Clinical Trials Directive poses 'a specific challenge [for] SME companies when developing new products for rare disorders that affect a limited number of patients. […] Typically, SMEs do not have in-house resources to track and manage national regulatory documentation, translations and approval processes.'[11]

           Non-commercial research: The Federation of the European Academies of Medicines, in its response to the 2009/10 public consultation highlighted that 'The Clinical Trials Directive has not solved the problems it was designed to do, but has dramatically increased administrative burden and costs for academia, resulting in a deterrent effect on new clinical research. […] In consequence of the Clinical Trials Directive, the EU has become a less attractive location of such research.'[12] The European Science Foundation (ESF), together with the European Medical Research Councils (EMRC), have highlighted that "current EU legislation represents a major hurdle to improving medical treatment due to the straight-jacket of EU legislation that the 2001 Clinical Trials Directive imposes".[13] The severe criticism of the Clinical Trials Directive has also led to a high number of academic publications painting a picture of increased bureaucracy and costs, accompanied by a reduction in important research activities.[14] These publications highlight the 'Regulatory impediments [which] jeopardize the conduct of clinical trials in Europe funded by the National Institutes of Health', [15] the 'Harmful impact of EU clinical trials directive',[16] leading to 'the death of academic clinical trials'.[17]

           Member States: The Heads of Medicines Agencies (HMA), an intergovernmental body bringing together the heads of all Medicines Agencies of the EU, has, in its 'Strategy for the Heads of Medicines Agencies 2011-2014'[18] called for "the creation of an efficient and unified regulatory environment for clinical trials in Europe that encourages innovation and high quality clinical research, by improving efficiency and reducing duplication, focussing assessment and inspections for clinical trials on a risk-based approach and promoting harmonised interpretation and implementation of guidelines and legislation related to clinical trials". This confirms the viewpoint of Member States who, in a statement made in Council in 2010, called upon the Commission to treat revision of the Clinical Trials Directive ‘as a matter of urgency'.[19] Apart from joint statements, there have been statements by individual Member States. For example the UK government, in its reply to the 2011 public consultation stressed that "the forthcoming review of the Directive provides an important opportunity to ensure that the EU maintains its position as an attractive place for the conduct of clinical trials necessary to the development of new medicines."[20]

           Union institutions: The European Parliament and the Council had also called repeatedly for revision of the Clinical Trials Directive. Examples include the Council Conclusions of 9 December 2010 on innovation and solidarity in pharmaceuticals, which call upon the Commission to ‘give priority to revising the Clinical Trials Directive’[21] and the European Parliament Resolution of 10 April 2008 on combating cancer in the enlarged European Union, which ‘calls on the Commission to revise [the Clinical Trials] Directive […] to encourage more academic research on cancer’.[22]

The criticism of the Clinical Trials Directive has also found its way into non-scientific publications stressing for example the bureaucracy created by the Clinical Trials Directive ('Les experts passent de plus en plus de temps à faire de la bureaucratie, aux dépens de la recherche clinique'),[23] its negative impact on public health ('British patients may be denied access to the latest drugs and treatments as a result of EU rules on clinical trials'),[24] and its negative impact on innovation and research ('EU Regulations hindering drug development, say charities').[25]

(4) Globally applicable principles of GCP

A range of internationally agreed documents set out universally applicable principles on protection of participants in clinical trials, no matter where the trial is performed. Studies suggest that, between 1947 and 2000, almost 400 international codes on the conduct of biomedical research have been adopted by various international bodies.[26] Of these, there are some key documents, such as:

· The revised version of the World Medical Association Declaration of Helsinki — Ethical Principles for Medical Research Involving Human Subjects;[27] and

· The Guideline E6 on Good Clinical Practice of the International Conference on Harmonisation of Technical Requirements for Registration of Pharmaceuticals for Human Use ('ICH').[28]

Moreover, at a more detailed level, international guidelines have been agreed on a variety of matters, such as the structure and content of clinical trial reports,[29] choice of control groups, statistical principles, etc.[30]

Finally, there are conventions on this matter which have been concluded under binding international law such as

· The Council of Europe (CoE) Convention for the Protection of Human Rights and Fundamental Freedoms;[31] and

· The CoE Convention on Human Rights and Biomedicine (‘Oviedo Convention’).[32]

(5) Sponsors involved in clinical trials

Clinical trials are performed under the responsibility of a sponsor. The sponsor is the person responsible for the trial. The notion of 'sponsoring' in this context must not be confused with the 'funding' of a clinical trial. The types of sponsors vary widely, from large multinational pharmaceutical companies and large research organisations with well-organised structures to small, fragmented cooperative structures with a far lower level of dedicated resources. However, these structures are often interlinked: for example, research organisations may carry out clinical trials for pharmaceutical companies and clinical research and their publications may influence the development of medicinal products.

(6) Authorisation by national competent authorities and Ethics Committees, inspections and surveillance

Clinical trials are subject to authorisation by the NCA and the EC of the Member State where the clinical trial is performed (the ‘Member State concerned’).

The Clinical Trials Directive is based on the concept of one EC opinion per Member State concerned. However, several Member States have a decentralised system where the single EC opinion is based on the opinions of several local committees. As a consequence, in the EU approximately 2 000 ECs are involved in assessment of clinical trials (see Annex 2).

Apart from this ex-ante control, regulatory compliance is verified by means of inspections of clinical trial sites by NCAs. According to information uploaded in EudraCT, since May 2004, approximately 3 150 inspections have been performed in the EU by NCAs. Clinical trials in non-EU countries are inspected only in the course of marketing authorisation procedures.

              Annex 2: Key figures

1.           Introduction

This annex sets out the key figures used in the calculations of the costs of the various policy options discussed.

Unless indicated otherwise, the sources are:

· The EU database for clinical trials ‘EudraCT’[33]; or

· The Agency report ‘Clinical trials submitted in marketing authorisation applications to the EMA — Overview of patient recruitment and the geographical location of investigator sites’.[34]

All figures on the duration of action or costs per man-hour were checked with stakeholders in the 2011 public consultation and the related workshops (see point Error! Reference source not found.).

2.           Number of clinical trials in the EU

Table 1: Clinical trials by year, by phase and by sponsor status[35]

|| Sponsor status || 2005 || 2006 || 2007 || 2008 || 2009 || 2010

Phase I || Commercial || 1 217 || 1 271 || 1 324 || 1 348 || 1 240 || 1 190

|| Non-commercial || 124 || 158 || 173 || 184 || 212 || 187

|| Unspecified || 9 || 8 || 6 || 4 || 4 || 7

Phase I total || 1 350 || 1 437 || 1 503 || 1 536 || 1 456 || 1 384

Phase II || Commercial || 622 || 647 || 806 || 696 || 685 || 597

|| Non-commercial || 310 || 489 || 682 || 601 || 663 || 585

|| Unspecified || 7 || 5 || 4 || 9 || 6 || 4

Phase II total || 939 || 1 141 || 1 492 || 1 306 || 1 354 || 1 186

Phase III || Commercial || 686 || 673 || 704 || 603 || 564 || 620

|| Non-commercial || 187 || 272 || 426 || 331 || 346 || 296

|| Unspecified || 6 || 6 || 6 || 3 || 4 || 1

Phase III total || 879 || 951 || 1 136 || 937 || 914 || 917

Phase IV || Commercial || 243 || 207 || 214 || 165 || 142 || 136

|| Non-commercial || 514 || 538 || 664 || 637 || 618 || 552

|| Unspecified || 7 || 11 || 7 || 4 || 13 || 19

Phase IV total || 764 || 756 || 885 || 806 || 773 || 707

Table 2: Number of clinical trials applied for in the EU per year since 2007[36]

2007 || 2008 || 2009 || 2010 || 2011

5 028 || 4'627 || 4 619 || 4'400 || 3490

Table 3: Number of applications to conduct clinical trials in the EU[37]

Sponsor status || 2005 || 2006 || 2007 || 2008 || 2009 || 2010

Commercial || 5 865 || 6 714 || 7686 || 7 993 || 7 655 || 7 672

Non-commercial || 1 303 || 1 677 || 2 216 || 2 039 || 2 161 || 2 037

Unspecified || 62 || 73 || 47 || 56 || 53 || 68

Total || 7 218 || 8 446 || 9 949 || 10 008 || 9 869 || 9 763

3.           Number of multinational clinical trials (EU)

Approximately 25 % of EU clinical trials are performed in more than one EU Member State (see Table 4). This is equal to approximately 60 % of all applications to conduct clinical trials in the Member States.

Table 4: Number of Member States/NCAs involved per clinical trial per year[38]

|| Number of Member States Involved

Year || 1 || 2 || 3 || 4 || 5 || 6 || 7 || 8 || 9 || 10 || 11 || 12 || 13 || 14 || 15 || 16 || 17 || 18 || 19 || 20 || 21 || 22 || 23 || 24 || 25

2005 || 2,972 || 280 || 196 || 115 || 95 || 71 || 73 || 50 || 31 || 23 || 24 || 16 || 6 || 7 || 7 || 3 || 5 || 1 || || || || || || ||

2006 || 3,292 || 274 || 162 || 121 || 101 || 86 || 74 || 53 || 31 || 33 || 24 || 14 || 22 || 12 || 12 || 5 || 5 || 5 || 4 || 2 || 2 || 1 || || 1 ||

2007 || 3,840 || 297 || 183 || 153 || 95 || 88 || 75 || 67 || 49 || 41 || 32 || 29 || 21 || 22 || 13 || 7 || 2 || 7 || 1 || 2 || 2 || 1 || || 1 ||

2008 || 3,520 || 259 || 175 || 130 || 102 || 96 || 67 || 58 || 49 || 33 || 32 || 32 || 18 || 13 || 11 || 6 || 6 || 6 || 2 || 4 || || 3 || 2 || 3 ||

2009 || 3,573 || 228 || 194 || 128 || 96 || 91 || 63 || 59 || 44 || 26 || 25 || 17 || 26 || 11 || 17 || 4 || 6 || 2 || 3 || 4 || 1 || || || || 1

2010 || 3,357 || 227 || 166 || 99 || 124 || 89 || 70 || 58 || 54 || 40 || 32 || 15 || 17 || 11 || 11 || 4 || 7 || 4 || 4 || 6 || 2 || 2 || || || 1

4.           Number of clinical trials in each Member State

Table 5: Clinical trials by year, by Member State and by phase[39]

|| Trial type: Human pharmacology (Phase I) || Trial type: First administration to humans || Trial type: Bioequivalence study || Trial type: Other || Trial type: Therapeutic exploratory (Phase II) || Trial type: Therapeutic confirmatory (Phase III) || Trial type: Therapeutic use (Phase IV)

YEAR: || 05 || 06 || 07 || 08 || 09 || 10 || 05 || 06 || 07 || 08 || 09 || 10 || 05 || 06 || 07 || 08 || 09 || 10 || 05 || 06 || 07 || 08 || 09 || 10 || 05 || 06 || 07 || 08 || 09 || 10 || 05 || 06 || 07 || 08 || 09 || 10 || 05 || 06 || 07 || 08 || 09 || 10

AT || 29 || 32 || 24 || 42 || 45 || 44 || 3 || 12 || 4 || 12 || 9 || 11 || 3 || 5 || 3 || 5 || 3 || 2 || 20 || 16 || 19 || 29 || 27 || 26 || 78 || 96 || 96 || 120 || 101 || 107 || 157 || 164 || 152 || 186 || 125 || 156                 || 53 || 42 || 54 || 64 || 38 || 52

BE || 134 || 140 || 136 || 147 || 134 || 148 || 35 || 39 || 35 || 54 || 44 || 47 || 10 || 9 || 5 || 5 || 8 || 6 || 71 || 59 || 66 || 63 || 66 || 74 || 166 || 160 || 166 || 182 || 175 || 138 || 257 || 217 || 203 || 221 || 216 || 222 || 63 || 64 || 67 || 54 || 50 || 47

BG || || || 14 || 24 || 18 || 33 || || || 1 || 1 || 1 || 4 || || || 11 || 14 || 9 || 23 || || || 2 || 2 || 6 || 7 || || || 14 || 61 || 48 || 45 || || || 22 || 126 || 91 || 96 || || || 0 || 10 || 9 || 8

CY || || 0 || 0 || 0 || 0 || 0 || || 0 || 0 || 0 || 0 || 0 || || 0 || 0 || 0 || 0 || 0 || || 0 || 1 || 0 || 0 || 0 || || 1 || 1 || 0 || 0 || 1 || || 5 || 2 || 3 || 2 || 0 || || 1 || 1 || 0 || 0 || 0

CZ || 21 || 31 || 37 || 41 || 37 || 39 || 1 || 5 || 3 || 3 || 12 || 8 || 25 || 31 || 33 || 29 || 25 || 28 || 11 || 11 || 12 || 12 || 7 || 13 || 76 || 84 || 116 || 93 || 128 || 91 || 157 || 220 || 210 || 190 || 175 || 204 || 22 || 23 || 26 || 26 || 18 || 28

DK || 21 || 39 || 26 || 32 || 17 || 20 || 6 || 11 || 11 || 12 || 6 || 5 || 1 || 4 || 1 || 0 || 2 || 1 || 18 || 24 || 15 || 24 || 13 || 17 || 74 || 94 || 79 || 96 || 96 || 70 || 111 || 146 || 126 || 149 || 120 || 97 || 57 || 55 || 41 || 60 || 41 || 44

EE || 3 || 2 || 2 || 1 || 0 || 0 || 1 || 1 || 0 || 1 || 0 || 1 || 0 || 0 || 0 || 1 || 0 || 0 || 1 || 1 || 2 || 0 || 1 || 1 || 20 || 22 || 23 || 16 || 23 || 14 || 57 || 59 || 65 || 59 || 47 || 59 || 7 || 5 || 4 || 7 || 7 || 3

FI || 17 || 31 || 36 || 54 || 28 || 22 || 3 || 3 || 6 || 6 || 1 || 4 || 8 || 8 || 8 || 12 || 6 || 6 || 11 || 27 || 27 || 34 || 20 || 17 || 39 || 61 || 52 || 53 || 45 || 33 || 109 || 132 || 115 || 111 || 77 || 102 || 34 || 59 || 47 || 50 || 53 || 39

FR || 53 || 122 || 232 || 219 || 215 || 200 || 7 || 29 || 53 || 39 || 60 || 51 || 5 || 10 || 24 || 15 || 11 || 16 || 29 || 79 || 157 || 148 || 126 || 123 || 98 || 175 || 338 || 286 || 346 || 313 || 183 || 281 || 389 || 344 || 383 || 388 || 39 || 57 || 118 || 80 || 98 || 88

DE || 311 || 477 || 356 || 372 || 435 || 367 || 56 || 101 || 50 || 48 || 56 || 56 || 48 || 73 || 46 || 79 || 69 || 52 || 174 || 245 || 164 || 150 || 216 || 192 || 280 || 417 || 414 || 421 || 460 || 395 || 374 || 506 || 456 || 447 || 464 || 448 || 122 || 165 || 119 || 126 || 103 || 102

EL || 0 || 33 || 17 || 3 || 11 || 11 || 0 || 1 || 1 || 0 || 1 || 0 || 0 || 33 || 13 || 2 || 5 || 2 || 0 || 7 || 4 || 1 || 5 || 8 || 0 || 31 || 34 || 11 || 43 || 19 || 2 || 96 || 87 || 28 || 140 || 56 || 0 || 16 || 19 || 10 || 25 || 10

HU || 15 || 17 || 29 || 50 || 35 || 33 || 2 || 3 || 4 || 10 || 4 || 4 || 9 || 14 || 18 || 28 || 22 || 18 || 6 || 13 || 8 || 10 || 11 || 12 || 61 || 77 || 95 || 99 || 106 || 85 || 132 || 173 || 197 || 174 || 154 || 186 || 14 || 23 || 28 || 25 || 18 || 23

IS || 0 || 2 || 2 || 1 || 0 || 0 || 0 || 1 || 1 || 1 || 0 || 0 || 0 || 1 || 0 || 0 || 0 || 0 || 2 || 1 || 0 || 0 || 0 || 0 || 3 || 6 || 0 || 2 || 6 || 6 || 11 || 7 || 6 || 9 || 3 || 5 || 2 || 4 || 2 || 1 || 2 || 1

IE || 11 || 11 || 24 || 20 || 21 || 14 || 5 || 6 || 1 || 2 || 1 || 1 || 9 || 13 || 19 || 15 || 16 || 8 || 2 || 3 || 4 || 8 || 3 || 4 || 20 || 23 || 19 || 27 || 18 || 20 || 50 || 54 || 44 || 49 || 47 || 43 || 16 || 29 || 19 || 27 || 12 || 13

             

             

|| Trial type: Human pharmacology (Phase I) || Trial type: First administration to humans || Trial type: Bioequivalence study || Trial type: Other || Trial type: Therapeutic exploratory (Phase II) || Trial type: Therapeutic confirmatory (Phase III) || Trial type: Therapeutic use (Phase IV

YEAR: || 05 || 06 || 07 || 08 || 09 || 10 || 05 || 06 || 07 || 08 || 09 || 10 || 05 || 06 || 07 || 08 || 09 || 10 || 05 || 06 || 07 || 08 || 09 || 10 || 05 || 06 || 07 || 08 || 09 || 10 || 05 || 06 || 07 || 08 || 09 || 10 || 05 || 06 || 07 || 08 || 09 || 10

IT || 8 || 12 || 22 || 74 || 83 || 59 || 0 || 0 || 0 || 19 || 22 || 19 || 6 || 13 || 18 || 11 || 8 || 6 || 1 || 0 || 6 || 28 || 19 || 24 || 145 || 208 || 529 || 326 || 369 || 245 || 213 || 280 || 603 || 431 || 371 || 289 || 35 || 58 || 152 || 111 || 161 || 94

LV || 3 || 3 || 2 || 3 || 2 || 2 || 1 || 0 || 1 || 2 || 2 || 0 || 2 || 0 || 0 || 0 || 0 || 1 || 0 || 2 || 5 || 1 || 0 || 2 || 15 || 20 || 32 || 27 || 28 || 18 || 42 || 51 || 56 || 58 || 45 || 57 || 7 || 2 || 4 || 9 || 1 || 2

LT || 0 || 3 || 0 || 1 || 1 || 0 || 0 || 0 || 0 || 1 || 1 || 0 || 0 || 1 || 0 || 0 || 0 || 0 || 1 || 1 || 1 || 0 || 0 || 0 || 22 || 28 || 42 || 30 || 23 || 17 || 60 || 75 || 79 || 75 || 51 || 77 || 9 || 8 || 4 || 7 || 2 || 2

LU || || || || 0 || || 0 || || || || 0 || || 0 || || || || 0 || || 0 || || || || 0 || || 0 || || || || 0 || || 1 || || || || 0 || || 7 || || || || 1 || || 0

MT || 0 || 0 || 0 || 0 || 0 || 0 || 0 || 0 || 0 || 0 || 0 || 0 || 0 || 0 || 0 || 0 || 0 || 0 || 0 || 0 || 0 || 0 || 0 || 0 || 0 || 1 || 0 || 2 || 0 || 0 || 1 || 3 || 3 || 3 || 4 || 3 || 0 || 0 || 2 || 0 || 0 || 1

NL || || 80 || 146 || 156 || 105 || 15 || || 21 || 45 || 51 || 41 || 4 || || 5 || 7 || 7 || 5 || 0 || || 57 || 101 || 95 || 59 || 9 || || 102 || 162 || 162 || 99 || 10 || || 126 || 175 || 214 || 140 || 31 || || 62 || 98 || 105 || 64 || 6

NO || 5 || 8 || 9 || 8 || 6 || 10 || 2 || 2 || 3 || 5 || 1 || 0 || 0 || 2 || 0 || 1 || 0 || 0 || 9 || 8 || 7 || 2 || 5 || 11 || 38 || 42 || 51 || 47 || 46 || 34 || 94 || 82 || 81 || 87 || 54 || 56 || 41 || 38 || 40 || 30 || 26 || 20

PL || || || 31 || 42 || 38 || 39 || || || 5 || 8 || 4 || 9 || || || 16 || 20 || 16 || 22 || || || 17 || 17 || 15 || 13 || || || 122 || 156 || 198 || 136 || || || 203 || 259 || 276 || 230 || || || 20 || 29 || 33 || 20

PO || 1 || 3 || 8 || 4 || 6 || 2 || 0 || 0 || 0 || 0 || 1 || 0 || 0 || 0 || 3 || 0 || 0 || 1 || 1 || 3 || 7 || 4 || 7 || 1 || 12 || 18 || 31 || 28 || 28 || 18 || 32 || 120 || 74 || 101 || 73 || 77 || 6 || 20 || 25 || 15 || 11 || 10

RO || || || || 0 || 5 || 9 || || || || 0 || 2 || 2 || || || || 0 || 0 || 2 || || || || 0 || 5 || 7 || || || || 9 || 38 || 82 || || || || 9 || 77 || 166 || || || || 2 || 5 || 14

SL || 7 || 5 || 1 || 15 || 9 || 11 || 1 || 0 || 0 || 1 || 0 || 1 || 2 || 2 || 1 || 4 || 5 || 4 || 2 || 4 || 1 || 9 || 5 || 8 || 24 || 23 || 15 || 63 || 35 || 37 || 53 || 91 || 39 || 160 || 73 || 113 || 5 || 8 || 4 || 13 || 7 || 10

SI || 0 || 1 || 1 || 0 || 1 || 1 || 0 || 1 || 1 || 0 || 1 || 0 || 4 || 2 || 0 || 0 || 0 || 0 || 0 || 0 || 1 || 0 || 1 || 0 || 9 || 5 || 4 || 7 || 7 || 9 || 19 || 19 || 20 || 15 || 12 || 11 || 2 || 8 || 6 || 4 || 8 || 6

ES || 87 || 76 || 82 || 101 || 110 || 159 || 9 || 10 || 15 || 18 || 22 || 36 || 40 || 44 || 33 || 35 || 42 || 55 || 42 || 24 || 31 || 52 || 47 || 63 || 138 || 150 || 194 || 204 || 248 || 269 || 241 || 262 || 261 || 275 || 265 || 396 || 104 || 100 || 90 || 124 || 100 || 142

SE || 77 || 91 || 98 || 81 || 72 || 57 || 20 || 22 || 31 || 26 || 18 || 10 || 8 || 10 || 2 || 6 || 4 || 2 || 40 || 61 || 52 || 51 || 49 || 50 || 89 || 116 || 117 || 109 || 111 || 74 || 154 || 162 || 171 || 166 || 144 || 139 || 69 || 52 || 58 || 53 || 56 || 39

UK || 625 || 333 || 345 || 317 || 282 || 392 || 154 || 78 || 100 || 83 || 76 || 93 || 57 || 34 || 35 || 26 || 21 || 31 || 346 || 170 || 168 || 183 || 134 || 201 || 387 || 266 || 294 || 275 || 298 || 344 || 429 || 248 || 280 || 307 || 253 || 386 || 341 || 198 || 202 || 189 || 150 || 205

Total || 1428 || 1552 || 1680 || 1808 || 1716 || 1687 || 306 || 346 || 371 || 403 || 386 || 366 || 237 || 314 || 296 || 315 || 277 || 286 || 787 || 816 || 878 || 923 || 847 || 883 || 1794 || 2226 || 3040 || 2912 || 3123 || 2631 || 2938 || 3579 || 4119 || 4256 || 3882 || 4100 || 1048 || 1097 || 1250 || 1232 || 1098 || 1029

             

5.           Number of subjects planned

Table 6: Number of subjects planned per year per clinical trial

MS concerned || 2007 || 2008 || 2009 || 2010

1 || 134,954 || 87,73 || 98,056 || 130,780

2 || 51,726 || 33,422 || 31,323 || 33,412

3 || 45,043 || 34,310 || 29,374 || 33,962

4 || 61,973 || 19,839 || 23,871 || 25,561

5 || 18,207 || 21,573 || 19,94 || 26,726

6 || 25,724 || 22,12 || 20,129 || 20,202

7 || 24,254 || 15,2300 || 12,308 || 15,638

8 || 16,705 || 16,346 || 11,948 || 11,992

9 || 12,98 || 38,838 || 12,308 || 28,467

10 || 18,974 || 8,343 || 11,455 || 15,909

11 || 13,525 || 11,000 || 10,519 || 4,280

12 || 11,928 || 11,585 || 8,118 || 3,139

13 || 12,401 || 11,552 || 16,849 || 3,985

14 || 13,023 || 4,526 || 7,818 || 10,383

15 || 21,970 || 6,949 || 11,083 || 2,435

16 || 16,231 || 13,956 || 2,200 || 5,744

17 || 1,000 || 4,193 || 4,824 || 3,200

18 || 7,843 || 11,173 || 1,307 || 7,775

19 || 850 || 9,062 || 8,300 || 3,715

20 || 10,665 || 13,229 || 8,331 || 5,143

21 || 5,900 || - || 1,815 || 6,340

22 || 4,395 || 7,507 || - || -

23 || - || 3,320 || - || -

24 || 4,500 || 2,817 || 8,000 || 743

25 || - || - || - || -

26 || - || - || - || -

27 || -- || - || - || -

Table 7: Number of planned subjects in clinical trials in the EU[40]

Year || 2007 || 2008 || 2009 || 2010

|| 544 287 || 410 568 || 367 036 || 408 294

Table 8: Total number of planned subjects in clinical trials with at least one site in the EU

Year || 2007 || 2008 || 2009 || 2010

|| 1 043 642 || 781 695 || 677 723 || 881 546

             

6.           Number of substantial amendments

1.1. Submissions to NCAs

As a rough estimate, every year each clinical trial is amended, on average, twice, insofar as submission to the NCA is concerned. This means that each year approximately 24 000 SAs are made to clinical trials in the EU as far as this is of relevance for NCAs.

This is confirmed by a survey of the Commission amongst Member States (Table 9).

Table 9: Number of substantial amendments submitted to NCAs (2010)[41]

AT || 945[42]

BE || 1610

BG || 489

CY ||

CZ || 1880 including  IPS/ICF

DE (BfArM) || 4256

DE (PEI) || 914

DK || 609 (Jan-Nov)

EE || 180

EL || 362

ES || 5290 SA received, out of which we estimate  2973 were for assessment

FI || 358

FR || 3166 : 1609 for authorisation + 1557 for information

HU || 1604

IE || 399

IT || 165[43]

LT || 254

LU ||

LV || 190

MT || 14

NL || 1401

PL || 1100

PT || 260

RO || 1367

SE || 892

SI || 66

SK ||

UK || 4563

IS || 18

This estimate is further supported by ICREL: every year, approximately 1 000 SAs are submitted, on average, per Member State,[44] i.e. approximately 27 000 SAs per year.[45]

1.2. Submissions to ECs

There are no precise figures for SAs submitted to ECs per Member State. It can be assumed that the number is higher than for NCAs, because most Member States consider adding further investigators or trial centres as a SA. Both investigators and trial centres change frequently. Therefore, it is realistic to assume that the number of SAs submitted to ECs is twice the number submitted to NCAs, i.e. 2 000 SAs per Member State, giving a total of 54 000 SAs.

1.3. Submissions per year (total)

Based on the above it can be assumed that each year 78 000 SAs are submitted to the ECs and NCAs in the EU.

7.           Number of SUSARs and SUSAR reports

Table 10: Number of SUSARs and SUSAR reports — Data from Eudravigilance – Clinical Trials Module (EVCTM)

|| 2009 || 2010

SUSARs || || 35 409

SUSAR reports || 94 600 || 99 583

Table 11: Number of SUSARs and SUSAR reports — Data from Member States (2009/2010)[46]

MS || Number of (unique) SUSARs (2009) || Number of SUSAR reports  (2009) || Number of (unique) SUSARs (2010) || Number of SUSAR reports (2010)

AT || N/A[47] || 178[48] || N/A || 382/657[49]

BE[50] || 980 (initial reports) || 6334 (initial reports and follow-up) || 785 (initial reports) || 3293 (initial reports and follow-up)

BG || 28 (initial reports) || 191 (initial reports and follow-up) || 36 (initial reports) || 160 (initial reports and follow-up)

CY || || || ||

CZ || 142 SUSAR reports, right No 49 SUSAR(only from CZ) || || 129 ||

DE (BfArM) || 23763 (3594  occurred in MS, 20169 outside MS); numbers of initial report without follow up. Note that it is not possible to count the number of SUSARs correctly by taking this number. || 60350 (9267 occurred in MS, 51083 outside MS); numbers of follow up reports without the initial report. . Note that it is not possible to count the number of SUSARs correctly by taking this number. || 23351 (3923 occurred in MS, 19428 outside MS); numbers of initial report without follow up. Note that it is not possible to count the number of SUSARs correctly by taking this number. || 64114 (9855 occurred in MS, 54259 outside MS); numbers of follow up reports without the initial report. . Note that it is not possible to count the number of SUSARs correctly by taking this number.

DE (PEI) || 1137 || 2527(initial reports and follow-up) || 1253 || 2684 (initial reports and follow-up)

DK || 133 national SUSARs || 291 incl. follow ups (National). 2848 national and EU SUSARs || || 200 (Jan-Nov) national SUSARs

EE || 19 || || 18 ||

EL || 306 || || ||

ES || NA || NA || NA || NA

FI || || 333 || || 290

FR || || 48 562 reports (initial + follow up/ local + outside) || || 48742 (initial + follow.up; local + outside)

HU || 238 || About 3-5 times as much as the unique SUSARs (it’s not possible to generate this data from our database) || 242 || About 3-5 times as much as the unique SUSARs (it’s not possible to generate this data from our database)

IE || 44 || N/A || 63 || N/A

IT[51] || 764 || 1632 || 706 || 1850

LT || NA || 301 (national SUSARs reports) || NA || 385 (national SUSARs reports)

LU || || || ||

LV || || 193 (SUSAR reports) || || 100 (national SUSAR reports)

MT || 0 || 0 || 3 (SUSAR reports) || 3 (SUSAR reports)

NL || || 1607 || || 1268

PL || || 17183 || ||

PT || 84 (national SUSAR) || 250 (national SUSAR reports) || 70 (national SUSAR) || 250 (national SUSAR reports)

RO || 256 || 328 || 79 || 191

SE || NA || 231 || NA || 238

SI || 16 national SUSARs || N/A || 17 national SUSARs || N/A

SK || || || ||

UK || 16 919 (1624 UK + 15295 foreign) || 33292 (2938 UK + 30354 foreign) || 19962 (1777 UK and 18185 foreign) || 49106 (4053 UK + 45053)

IS || 2 || 2 || 2 || 2

|| || || ||

The system for SUSAR reporting in the EU is highly diverse. It is therefore not surprising that the figures above diverge so widely, particularly when they are compared with the number of clinical trials performed in each Member State (see Table 5).

Based on these data, however, it can be assumed that each year approximately 35 000 SUSARs occur in the EU. This leads to approximately 200 000 SUSAR reports at national level (NCAs and ECs) and another 100 000 at EU level (phase IV centrally authorised medicines). The latter are not always reported by the sponsor but, depending on the Member State, might be reported by the NCA. It can be assumed that the sponsors submit approximately 50 % of these reports, i.e. 50 000 reports.

It can therefore be concluded that, every year, sponsors submit approximately 250 000 SUSAR reports to different bodies in the EU (Agency, EC, NCA).

8.           Number of annual safety reports ('ASR')

The number of ASR submissions equals the number of ongoing clinical trials in each Member State concerned.

The number of ASRs actually drafted is lower. This is for the following two reasons:

(a)      The Sponsor may submit an identical ASR for several clinical trials performed by that sponsor with the same IMP[52] (see table 12);

Table 12: Number of ASRs (2009/2010)[53]

MS || Number of ASRs (2009) || Number of ASRs (2010)

AT[54] || 713 || 757

BE[55] || 830 || 1 031

BG || 243 (containing all CTs with the IMP) || 221 (containing all CTs with the IMP)

CY || ||

CZ || N/A || not followed for this year, for next-yes

DE (BfArM) || 1287 This data are ASR submissions.  It cannot be specified if these ASR cover one or more than one CT or /and one ore more IMP. With these data it is neither possible to calculate ASR numbers of IMPs (now DSUR) nor ASR number of CTs. || 1850 This data are ASR submissions.  It cannot be specified if these ASR cover one or more than one CT or /and one ore more IMP. With these data it is neither possible to calculate ASR numbers of IMPs (now DSUR) nor ASR number of CTs. 

DE (PEI) || 247(containing all CT with the IMP) || 340 (containing all CT with the IMP)

DK || 494 || 399 (Jan-Nov)

EE || N/A || N/A

EL || 221 ||

ES || Estimated 1 440, not checked if they are or not unique || 1288 ASR received in 2011, not checked if they are unique

FI || 457 || 425

FR || 1 070 || 1 095

HU || N/A || 1074

IE || 144 || 158

IT[56] || 1202 || 1111

LT[57] || 200 || 200

LU || ||

LV || 80 || 114

MT || N/A || 8

NL || 441 || 432

PL || 477 || 527

PT || 160 || 179

RO || 426 ‘received reports’ || 567 ‘received reports’

SE || 721 || 699

SI || 72 || 76

SK || ||

UK || 1 568 || 1 756

IS || 7 || 12

(b)     Moreover, the figure of total ASR actually drafted as contained in table 12 has to be reduced further because ASRs submitted to different Member States may be identical.[58]

Therefore, to establish the number of ASRs drafted one has to refer to the number of IMPs involved in clinical trials per sponsor.

There is no reliable hard data available on the number of IMPs involved in clinical trials performed at a given point in time in the EU per sponsor. However, it can be assumed from what is mentioned above that the 11 000 ASR versions received by Member States involve approx. 70% of IMPs per sponsor (a sponsor may conduct several clinical trials with the same IMP). This means that there are at a given moment approximately 7 700 IMPs on a per-sponsor basis. This means that each year approximately 7 700 ASRs are drafted by sponsors conducting clinical trials in the EU.

9.           Costs of clinical trials

The costs of clinical trials performed in Europe are as follows:

As regards industry-driven research, in 2008 the pharmaceutical industry invested about 26 000 m EUR in research and development,[59] of which 68 % were allocated to clinical trials, i.e. 15 600 m EUR (a).

No figures are available for trials other than industry-driven clinical trials. On the basis of industry driven research investment allocated to clinical trials and taking into account that forty per cent (see Table 1) of all clinical trials are performed by ‘non-commercial sponsors’ it could be argued that the investment by ‘non-commercial sponsors’ is:

4/6*a = 10.4 m EUR = b.

It has to be borne in mind, though, that clinical trials performed by ‘non-commercial sponsors’ tend to be less costly. For example, the IMPs used are often authorised and thus do not have specific distribution channels and profit from simplified labelling. Therefore, it is appropriate to deduct 30 % (c) of the costs compared with industry-driven research.

The total costs for clinical trials in Europe per year are therefore:

a+b*0.7 = 22 880 m EUR.

These costs are mainly generated by the activities listed below. While the costs depend very much on a case-by-case basis[60], an attempt has been made, in meeting with stakeholders,[61] to rank them in importance in terms of costs:

· Quality assurance during conduct of trial: Communicating with clinical trial centres, including monitoring and surveillance (staff costs and support services, such as translation, travel, accommodation, couriers, etc.);

· Remunerating sites and investigators (incl. possible trainings);

· Designing and drawing up the protocol;

· Preparing (including manufacturing and blinding) or purchasing the IMPs[62];

· Distributing the IMPs to the clinical trial centres;

· Analysing data (incl. Data Safety Monitoring Board);

· Administrative costs;

· Insurance;

· Fees.

10.         Share of SMEs, including micro-enterprises

Costs for the conduct of clinical trials are ultimately born by the sponsor. Where costs created by regulation (administrative costs or other compliance costs) fall on the investigator (or the respective clinical trial site), they are usually passed on to the sponsor by way of contractual arrangements (see above, 'remuneration of sites and investigators').

In 2010, the share of clinical trials was as follows:

· 1 620 clinical trials were sponsored by 'academic sponsors' (a);

· 2 543 clinical trials were sponsored by 'commercial sponsors', i.e. pharmaceutical companies (b).

Amongst academic sponsors, it can be assumed that practically none falls within the definition of a SME: According to the EU definition in Commission Recommendation 2003/361/EC[63] an enterprise is considered to be any entity engaged in an economic activity. Moreover, Article 4(3) of the Annex to that Recommendation provides that an enterprise cannot be considered an SME if 25% or more of the capital or voting rights are directly or indirectly controlled, jointly or individually, by one or more public bodies.

Amongst pharmaceutical companies, it is assumed that approximately 40% fall within the SME definition of the EU.[64] However, a large part of these pharmaceutical companies are active in the area of generic and/or over-the-counter medicines. In these sectors clinical trials activity is rather limited. In particular, pharmaceutical companies for generic medicines limit their activity largely to bioequivalence studies and pharmacokinetic studies. Therefore, it can be assumed that the by far larger part (approximately 85%) of the 2 543 clinical trials of 'commercial studies' are conducted by the 60% of pharmaceutical companies not falling within the EU-definition of an SME. The share of pharmaceutical SMEs sponsoring clinical trials is thus approximately 15%.

The share of SMEs who have to bear the costs for clinical trials is thus as follows:

(b*0.15)/(a+b) = 0.09 = 9%

Micro-enterprises are the smallest category of SME, with less than ten employees and a turnover or balance sheet total equal to or less than 2m EUR.[65]

In view of the complexities of the regulatory and business environment in the pharmaceutical sector, in particular in the area of clinical research, it can be assumed that practically no micro-enterprise is active as sponsor in the area of clinical trials.

11.         Staff figures in national competent authorities

Table 13: Number of staff in NCAs[66]

|| || Clinical trial assessment || Assessing of safety reports

|| MS || Quality assessors || Non clinical assessors || Clinical assessors || External experts || Validation assessors || Other assessors (details + number) || Number of SUSARs assessors || Number of ASR assessors || Other (detailed + number) ||

|| AT || 1.5 || 1 || 2 || none || 1.75 || || 0.25 || 0.5 || ||

|| BE || 1.5 || 1.5 || 0.5 || When appropriate in light of expertise || 3 || || 0 || 0 || 0 ||

|| BG || || || || 2 part time clinical assessors || 1 || || 0* || 0* || ||

|| CY || || || || || || || || || ||

|| CZ || 2 full time; 3x0.75 part time; 2x0.2 || 3x0.2 || 3x0.75; 2 full time || Flexible; according to our demand || 1 || Medical device according to state, other assessors are independent of SUKL( Ministry of Envir.- GMO, SÚJB-state office for nuclear safety –radiopharma.IMP) || 0 Assessment by clinical assessors according to their indication ||  0 || ||

|| DE (BfArM) || || || || || || || || || ||

|| DE (PEI) || Figures of the past were an estimation that is not considered reliable enough to be used or published. In order to provide reliable figures an in depth analysis of each Clinical Trial Application and Approval would be required. ||

|| DK || Not in clinical trial department, but borrows from licensing department (risk based) || 1 || See external experts || 5 part time clinical assessors (10-15 h/week) – 1,5 FTE || 1 || 2,5 pharmacists 1,5 administrative assistants || Part-time allocation so SUSARs app. 10 h/week (incl. EVCTM) ~ 0.25 || Part time allocation of clinical assessors ~ 0.5 || 0.5 for administrative handling ||

|| EE || Assessor from licensing department || 0 || 1 || As occasion requires. || Responsibility of clinical assessor and partly job of secretary || || 0 (clinical assessor's responsibility) || 0 (clinical assessor's responsibility) || ||

|| EL || 0 || 0 || 0 || As occasion requires || 2 || || 0 || 0 || 0 ||

|| ES Data expressed in FET || 1.25 for Chemical IMP 3 for biological and advanced therapy IMP || 1 || 4 || 1 || 3.5 (3 administrative supported by 1 pharmacist) || 2.5 pharmacist and 2 administrative in charge of database quality control and administrative procedures of the applications || 1* *Dedicated to safety issues of CT || 1* *Dedicated to safety issues of CT || ||

|| FI || From other departments, when needed. || From other departments, when needed. || From other departments, when needed. || On demand. || 0 || FIMEA clinical trials department: 1.5 permanent assessors for all clinical trial related tasks. || See column 'other assessors' || See column 'other assessors' || ||

|| FR (2011) ||  From other departments, when needed. ||  2,3 || 6.8 || Yes on demand ||  4,7 || 0.8 reg. affairs || 1.6 ||  1 || 0 ||

|| HU || 5 || 1 || 4 || 7 clinical assessors are available per demand || 0.5 || || 0.5 || 0 || 0 ||

|| IE || 0.9 FTE for biological products and 1FTE for chemical products || 0.5 || 1 FTE = (7 assessors  are available per demand from  the authorization and registration department) || on demand- External Experts participate in a monthly Clinical Trials Subcommittee where the trials are reviewed and are available on demand || 0 || 1 scientific officer (part-time) responsible for EudraCT. || 0.5 || Variable – Done by CTA assessors || ||

|| IT || 0.5 || 0.5 || 0.5 || Yes, on demand || 2 || || 0.33 || 0.33 || 0.33 ||

|| || || || || || || || || || ||

|| LT || 0,125 || 0,25 || 1,25 || On demand || 1 || - || 0 || 0 || ||

|| LU || || || || || || || || || ||

|| LV || 0,25 || 0,25 || 1,25 || Flexible;-quality and clinical experts are available on necessity || 0,25 || || 0 (done by CTA assessor) || 0 (done by CTA assessor) || ||

|| MT || 6 QAs shared with other procedures such as DCPs/MRPs || 1 shared with other procedures such as DCPs/MRPs || 3 shared with other procedures such as DCPs/MRPs || On demand || 6 pharmacists shared with other procedures || || 1 pharmacist shared with for all ICSR revievwed || 1 pharmacist shared with for all ICSR revievwed || ||

|| NL || || || || || || || 0 || 0 || / ||

|| PL || 2 || 0 || 0 || Yes: 2 non clinical and 20 clinical || 8 || || 0* || 0* || ||

|| PT || 0,5 (5 assessors shared with authoriz./ reg. department) || 0,5 (2 assessors shared with authoriz./ reg. department) || See external experts || 1 (6 clinical + 4 quality assessors available on demand) || 2,5 || || 0,1 (done by CTA assessor ) ||  0,1 (done by CTA assessor ) || 0,5 data entry, administrative handling and scientific review ||

|| RO || -2 part time for Chemical IMP -for biological IMP from other department when needed || 0 || 1 clinical assessor full time, 3 part time || 0 || 0 dedicated staff, job done by clinical assesorrs || 0 || 0 (only administrative handling) || 0 dedicated staff, job done by clinical assesorrs || 0 ||

|| SE || 2 || 2 || 4 || 0.7 || 0.5 || - || 0.20 || 1.0 || Total number of SUSAR reports per year increased 2011 (prel number as of mid-Dec = 281) ||

|| SI || 0 || 0 || 0.5 || 4 || 1 || || 0 || 0.1 || ||

|| SK || || || || || || || || || ||

|| UK || 4 || 2.5 || 6 || Expert Advisory panel when required. || 0 || 2 scientific assessors (safety); 1 scientific assessor (amendments) || 1.25 || 1.25 || 2 Scientific assessors (safety) ||

IS || Assessors from licencing department as needed || Access to 5 assessors as needed || Access to 5 clinical assessors as needed || Access to statistician and toxicologist as needed || 3 administrators as needed || N/A || Done by clinical assessors || Done by clinical assessors || Quality/other assessors when necessary ||

12.         Number of Ethics Committees

Table 14: Number of Ethics Committees in EU Member States[67]

|| Number of Ethics Committees || Number of Ethics Committees (including local ethics committees)

Austria || 27 ||

Belgium || 35 || 215

Bulgaria || 103 ||

Czech || 9 || >100

Cyprus || 1 ||

Denmark || 8 ||

Estonia || 2 ||

Finland || 25 ||

France || 40 ||

Germany || 53 ||

Greece || 1 ||

Hungary || 1 ||

Ireland || 13 || 40

Italy || 264 || >900

Latvia || 5 ||

Lithuania || 2 ||

Luxembourg || 1 ||

Malta || 1 ||

Netherlands || 31 ||

Poland || 55 ||

Portugal || 1 ||

Romania || 1 ||

Slovakia || 9 || 89

Slovenia || 1 ||

Spain || 136 ||

Sweden || 8 ||

UK || 126 ||

             

13.         Cost per man-hour

One man-hour for work on regulatory affairs relating to clinical trials costs approximately 60 EUR.[68] This number exceeds the average EU tariff used in particular for calculation of administrative costs by the Commission. This can be explained by the relatively high salaries in the sector of pharmaceutical research and regulatory affairs.

The figure has been double-checked and confirmed with stakeholders at various occasions, including in the 2011 public consultation.

14.         Duration of a clinical trial

In terms of clinical trial regulation, the duration of a clinical trial starts with the first authorisation of a clinical trial in a Member State in the EU, and ends with the 'end of the trial'. The end of the trial is defined by the sponsor in the protocol. Typically, it is the last visit of the last subject.

The duration of clinical trials vary greatly. They last from a few days or weeks to several years or even decades. The duration of a clinical trial depends inter alia on the type of clinical trial (phase I-IV), the subject population, and the endpoint.

Typically, today, clinical trials tend to last longer than in the past in view of the complexity of the design, the higher recruitment targets, and the choice of the endpoints.

While there is no hard data available, in view of the above considerations, it can be assumed that the average duration of a clinical trial in the EU is 3 years.

Table 14: Number of patients in pivotal trials submitted in MAAs to the EMA per region and year

No of patients || 2005 || 2006 || 2007 || 2008 || 2009 || Total

|| Σ || % || Σ || % || Σ || % || Σ || % || Σ || % || Σ || %

EU/EEA/EFTA || 32 090 || 37.0 || 49 960 || 44.2 || 55 667 || 44.1 || 42 024 || 28.6 || 51 628 || 42.1 || 231 369 || 38.8

comprising: || || || || || || || || || || || ||

EU-15/EEA || 27 822 || 32.1 || 30 714 || 27.2 || 42 894 || 34.0 || 27 561 || 18.7 || 33 711 || 27.5 || 162 702 || 27.3

EU-10 || 3 412 || 3.9 || 16 601 || 14.7 || 11 016 || 8.7 || 11 706 || 8.0 || 14 768 || 12.0 || 57 503 || 9.7

EU-2 || 656 || 0.8 || 2 146 || 1.9 || 1 251 || 1.0 || 2 447 || 1.7 || 2 628 || 2.1 || 9 128 || 1.5

Switzerland || 200 || 0.2 || 499 || 0.4 || 506 || 0.4 || 310 || 0.2 || 521 || 0.4 || 2 036 || 0.3

North America || 37 117 || 42.8 || 33 389 || 29.6 || 41 810 || 33.2 || 55 165 || 37.5 || 42 269 || 34.5 || 209 750 || 35.2

comprising: || || || || || || || || || || || ||

Canada || 3 477 || 4.0 || 3 919 || 3.5 || 6 231 || 4.9 || 4 454 || 3.0 || 9 581 || 7.8 || 27 662 || 4.6

USA || 33 640 || 38.8 || 29 470 || 26.1 || 35 579 || 28.2 || 50 711 || 34.5 || 32 688 || 26.7 || 182 088 || 30.6

Rest of world || 17 585 || 20.3 || 29 637 || 26.2 || 28 628 || 22.7 || 49 948 || 33.9 || 28 663 || 23.4 || 154 461 || 25.9

comprising: || || || || || || || || || || || ||

Africa || 523 || 0.6 || 1 938 || 1.7 || 2 061 || 1.6 || 9 962 || 6.8 || 3 431 || 2.8 || 17 915 || 3.0

Middle East/ Asia/Pacific || 1 694 || 2.0 || 9 925 || 8.8 || 7 801 || 6.2 || 17 458 || 11.9 || 9 627 || 7.9 || 46 505 || 7.8

Australia/ New Zealand || 1 560 || 1.8 || 1 892 || 1.7 || 2 663 || 2.1 || 1 219 || 0.8 || 1 344 || 1.1 || 8 678 || 1.5

CIS || 664 || 0.8 || 6 939 || 6.1 || 2 731 || 2.2 || 6 677 || 4.5 || 5 653 || 4.6 || 22 664 || 3.8

Eastern Europe (non-EU) || 69 || 0.1 || 862 || 0.8 || 1 202 || 1.0 || 1 370 || 0.9 || 539 || 0.4 || 4 042 || 0.7

Central/ South America || 13 075 || 15.1 || 8 081 || 7.2 || 12 170 || 9.7 || 13 262 || 9.0 || 8 069 || 6.6 || 54 657 || 9.2

Total || 86 792 || 100 || 112 986 || 100 || 126 105 || 100 || 147 137 || 100 || 122 560 || 100 || 595 580       || 100

             

Table 15: Inspections by the EMA in the EU and in non-EU countries

              Annex 3: Objective No 1 — A modern regulatory framework for submission, assessment and regulatory follow-up of applications for clinical trials

1.           Policy option No 1/1 (baseline)

1.4. Administrative costs per year

Overview

|| Number of actions || Duration in man-hours || Cost per man-hour for regulatory affairs || Total (in EUR) || Comments

Initial application || || || 60 EUR || ||

NCA || 9 763 || 40 || 23 431 200 ||

EC || 97 630 || 32 || 187 449 600 ||

Follow-up information || || || ||

NCA || 7 810 || 16 || 7 497 600 ||

EC || 7 810 || 16 || 7 497 600 ||

Substantial amendments || || || ||

NCA || 27 000 || 10 || 16 200 000 ||

EC || 54 000 || 10 || 32 400 000 ||

SUSAR reporting || || || ||

NCA || 250 000 || 1.5 || 22 500 000 ||

EC || 35 409 || 1.5 || 3 186 810 ||

ASR || || || ||

NCA || 29 289 || 1.5 || 2 636 010 ||

EC || 29 289 || 1.5 || 2 636 010 ||

End-of-trial reporting || || || ||

NCA || 9 763 || 0.5 || 292 890 ||

EC || 9 763 || 0.5 || 292 890 ||

TOTAL || 306 020 610 ||

Explanations

· General remark: All these administrative costs are 'recurring administrative costs' in the context of conducting clinical trials in the EU.

· Initial application:

o Number of dossiers handled (NCA): EudraCT delivers very precise figures (see Annex 2, point 2). This figure is based on 2010.

o Number of actions (EC): In practically every Member State submission to the EC is separate from submission to the NCA. Moreover, despite the fact that a ‘single opinion’ is given (see Annex 1), in many Member States submissions have to be sent to a multitude of ECs (for the number of ECs, see Annex 2, point 12). This figure is based on an estimate of submission, on average, to 10 ECs (a) for each of the 9 763 applications for clinical trials (b) in the Member States: a*b = 97 630.

o Duration per dossier handled: 5 man-days, i.e. 40 man-hours per application, for the NCA (exclusive preparation of study documents, the protocol, IMP dossier, investigator’s brochure, etc.) and 4 man-days, i.e. 32 man-hours per application, for the EC. This figure is based on data submitted in the two public consultations[69] and collated for the ‘EU project on baseline and reduction of administrative costs — Measurement data and analysis for the pharmaceuticals legislation priority area’. This figure takes into account that:

§ Submission to additional Member States is less costly than the initial submission;[70]

§ Dossiers have diverging degrees of complexity, depending on the type of clinical trial; and

§ The documentation to be submitted to ECs is usually lighter than the documentation to be submitted to the NCA (less information related to the IMP).[71]

· Follow-up information

o Number of actions (NCA): According to estimates by stakeholders, in approximately 80 % (a) of all applications to conduct a clinical trial an NCA requests additional information or raises grounds for non-acceptance: a*9 763 = 7 810.

o Number of actions (EC): The same holds true for follow-up information requested by ECs, i.e. 80 % (a) of all applications. However, in most Member States such requests are channelled, as the ‘single opinion’, via an EC. Therefore, the number of applications equals the number of NCAs, i.e. 9 763 (b). The number of dossiers handled is therefore: a*b = 7 810.

o Duration per action: Collecting and submitting this additional information takes, on average, approximately 2 man-days, i.e. 16 man-hours.[72]

· Substantial amendments (SAs)

o Number of actions (NCA and EC): See Annex 2, point 6.

o Duration per action: According to estimates by stakeholders, preparation and submission of an SA takes, on average, approximately 10 man-hours.[73]

· SUSAR reporting

o Number of actions (EC): Unlike NCAs (see Annex 2), the figures for SUSARs submitted to ECs are less certain. Many Member States have transposed the Clinical Trials Directive in such a way as to reduce the number of SUSARs reported to ECs. It can be assumed that an adverse reaction is, on average, reported only once to an EC, usually to the EC responsible in the Member State where the adverse reaction occurred. For this figure, see Annex 2.

o Duration per action: The time needed to submit the information related to SUSARs is approximately 1.5 man-hours.

· Annual safety report

o Number of actions (EC and NCA): The duration of a clinical trial is, on average, approximately 3 years (a). The 9 763 (b) applications mean that a*b = 29 289 ASRs have to be submitted. This holds true for NCAs and ECs. This number is independent of the fact that the number of actual ASRs is lower than the number of submitted ASRs (see Annex 2). As the ASR builds on the IMP, fewer ASRs are drafted than submitted (sponsors submit copies of an identical ASR).

o Duration per action: The time needed to submit the information related to ASRs is approximately 1.5 man-hours.

· End-of-trial declaration

o Number of actions (EC and NCA): The duration of a clinical trial is, on average, approximately 3 years. This means that 3 x 9 763 clinical trials ('in terms of applications) are ongoing at any given time. Of these, one third finish each year.

o Duration per action: The time needed to submit the information related to ASRs is approximately 0.5 man-hours.

1.5. Administrative burden

These administrative costs are to large extent administrative burdens. Most of these obligations to collect, process and report would not be performed if they were not provided for in the Clinical Trials Directive:

· Initial application: While some of the information would be gathered and processed, this information would not undergo a submission procedure, and certainly not a multiple submission procedure as provided for in the Clinical Trials Directive.

· Follow-up information: The administrative costs related to follow-up information would not be incurred if there was not the legal/regulatory requirement to provide such information.

· Substantial amendments: The same applies as for the initial application.

· SUSAR reporting: Sponsors would collect and process this information. However, unless this is provided in the legislation, they would not submit it to supervising authorities.

· Annual safety reporting: This information would not be collected, processed and submitted if this was not provided for by the legislation.

· End of trial reporting: This information would not be collected, processed and submitted if it was not provided for by the legislation.

In summary, and in view of the above explanations, one can conclude the following:

Overview:

Action || Administrative costs (in EUR) || Administrative burdens (as share of administrative costs) || Administrative burdens (in EUR)

Initial application || 210 880 800 || 80% || 168 704 640

Follow-up information || 14 995 200 || 100% || 14 995 200

Substantial amendments || 48 600 000 || 80% || 38 880 000

SUSAR reporting || 25 686 810 || 80% || 20 549 448

ASR || 5 272 020 || 100% || 5 272 020

End-of-trial reporting || 585 780 || 100% || 585 780

|| 306 020 610 = (a) || || 248 987 088 = (b)

It can be concluded that the share of administrative burden of all costs is as follows:

b/a = 0.814 = 81%

1.6. Other compliance costs

Apart from administrative costs, the Clinical Trials Directive also creates other compliance costs. These include:

· Preparing (including manufacturing and blinding) or purchasing the IMPs;

· Communicating with clinical trial centres, including monitoring and surveillance;[74]

· Analysing data;

· Insurance;

· Fees.

On the basis of publicly-available information[75] the Commission has held, from 2008 until 2011, discussions on the costs of clinical trials.[76]

While it is relativey straightforward to establish the total costs of clinical trials for sponsors (see Annex 2, point 9), and to establish the administrative costs and administrative burdens (see point 1.2 of this Annex), it is a challenge for sponsors to establish precisely which non-administrative costs are the results of a regulatory obligation (i.e. fall under the definition of 'other compliance costs') and which costs incur for other reasons, such as as standard or good practices of the organisation.

Despite these difficulties, in the abovementioned discussions it became clear that only a relatively minor part of non-administrative costs is actually a result of regulation. In particular, the costs for quality assurance during the conduct of the clinical trial (see Annex 2, point 9), which create in most cases the bulk of the costs for a clinical trial, are not directly caused by regulation. Rather, these costs follow from the inherent need to produce reliable and robust results, in order for the sponsor to have reasonable certainty that the data is not rejected or put otherwise in question.

A similar reasoning applies to the costs incurred for the designing and drawing up of the protocol: While it is a regulatory requirement to have a protocol for each clinical trial, the costs for designing the trial are not a direct consequence of regulation, but rather caused by the sponsor's interest to have a sound, reliable protocol which is going to address the question addressed in the clinical trial.

On the basis of these discussions a careful estimation allows for assuming that approximately 10 % of the costs for clinical trials in the EU fall under the definition of 'other compliance costs'. These totals thus approximately 22 000 m EUR (see Annex 2). Other compliance costs are therefore 2 200 m EUR per year.

2.           Policy option No 1/2 — Central submission with separate assessment

1.7. Administrative costs per year

The impact on administrative costs for initial submission would be as follows:

Overview

|| Number of actions || Duration in man-hours || Cost per man-hour for regulatory affairs || Total (in EUR) || Comments

Initial application || 4 400 || 40 || 60 EUR || 10 560 000 ||

Follow-up information || || || ||

NCA || 7 810 || 16 || 7 497 600 ||

EC || 7 810 || 16 || 7 497 600 ||

Substantial amendments || 24 000 || 10 || 14 400 000 ||

SUSAR reporting || 52 500 || 1.5 || 4 725 000 ||

ASR || 7 700 || 1.5 || 693 000 ||

End-of-trial reporting || 4 400 || 0.5 || 132 000 ||

TOTAL || 45 505 200 ||

Explanations

· Initial application: In this policy option, there would be one application per clinical trial.

· Follow-up information: In this policy option, the follow-up information would be dealt with as in the baseline option.

· Substantial amendments: On average, a clinical trial is amended approximately twice (a) per year. This includes SAs regarding trial sites, which are typically assessed by ECs. At any given time, there are approximately 12 000 clinical trials ongoing in the EU (b). The number of SAs submitted in this policy option would therefore be: a*b = 24 000.

· SUSAR reporting: As indicated in Annex 2, approximately 35 000 SUSARs (a) occur in the EU every year. In this policy option, each SUSAR would be reported only once. However, in view of possible follow-up reports, the number should be increased by 50 % (b): a+(a*b) = 52 500.

· Annual safety reporting: The report would be submitted only once per sponsor per IMP. Approximately 7 700 ASRs are drawn up each year (see Annex 2)

· End-of-trial reporting: On average, one third of all clinical trials in progress (4 400 in 2010) finish in any given year. A clinical trial lasts, on average, 3 years.

1.8. Implementation costs

Regarding the implementation costs for the Agency/Commission, reference is made to Annex 6.

3.           Policy option No 1/3 — Central submission with joint assessment

1.9. Administrative costs per year

Overview

|| Number of actions || Duration in man-hours || Cost per man-hour for regulatory affairs || Total (in EUR) || Comments

Initial application || 4 400 || 40 || 60 EUR || 10 560 000 ||

Follow-up information || 3 960 || 16 || 3 801 600 ||

Substantial amendments || 24 000 || 10 || 14 400 000 ||

SUSAR reporting || 52 500 || 1.5 || 4 725 000 ||

ASR || 7 700 || 1.5 || 693 000 ||

End-of-trial reporting || 4 149 || 0.5 || 132 000 ||

TOTAL || 34 300 600 ||

Explanations

The impact on administrative costs is identical to policy option No 1/2, with the exception of the follow-up information. It is assumed that this information is requested for 90 % (a) of all 4 400 clinical trials (b). This is in line with the baseline, where it is assumed that 80 % of all applications are followed up with questions: a*b = 3 960.

1.10. Other compliance costs

The other compliance costs generated by the Clinical Trials Directive are mainly linked to:

· Preparing (including manufacturing and relabelling) or purchasing the IMPs;

· Communicating with clinical trial centres, including monitoring and surveillance;

· Analysing data;

· Insurance;

· Fees.

The joint assessment proposed under this policy option would allow a common approach to the issues related to the preparation (incl. blinding.) of the IMP, monitoring and surveillance.

Currently approaches diverge between Member States, which adds to the costs for compliance with the Clinical Trials Directive.

In this context, tt is not possible to assess the costs/savings with the same degree of precision as for administrative costs. However, this matter was discussed with stakeholder experts during the various meetings and workshops between 2008 and 2011 (see point Error! Reference source not found.). On the basis of these discussions one can reasonably expect that the savings in other compliance costs under this policy option add up to 20 % of the other compliance costs in the baseline option, i.e. 0.2*2 200 m EUR = 440 m EUR. This estimation is based on

· A single product file, thus not requiring adaptation of the product charateristics to different Member States; and

· A single set of rules for analysing data and communicating between the clinical trial centres and the sponsors, thus not requiring varying standard operating procedures, with corresponding training and staff needs.

1.11. Implementation costs

Implementation costs at EU level:

This is set out in detail in Annex 7.

Implementation costs for national administrations

Costs for national administrations would go down insofar as joint assessment/mutual recognition would not necessarily require indepth assessment of the dossier by every Member State concerned.

4.           Policy option No 1/4 ‑ Central submission with central assessment

1.12. Administrative costs per year

The savings in terms of the submission procedure and follow-up would be similar to those generated by policy option No 1/3. However, additional costs would be generated by the ‘dual approval’, due to national follow-up questions.

|| Number of actions || Duration in man-hours || Cost per man-hour for regulatory affairs || Total (in EUR) || Comments

Initial application || 4 400 || 40 || 60 EUR || 10 560 000 ||

Follow-up information || || || ||

EMA level || 3 960 || 16 || 3 801 600 ||

National level || 7 810 || 16 || 7 497 600 ||

Substantial amendments || 24 000 || 10 || 14 400 000 ||

SUSAR reporting || 52 500 || 1.5 || 4 725 000 ||

ASR || 7 700 || 1.5 || 693 000 ||

End-of-trial reporting || 4 149 || 0.5 || 132 000 ||

TOTAL || 41 798 200 ||

Explanations — follow-up information

At EMA level, it is assumed that this information is requested for 90 % (a) of all 4 400 clinical trials (b). This is in line with the baseline, where it is assumed that 80 % of all applications are followed up with questions: a*b = 3 960.

At national level, the same reasoning as for policy option No 1/1 applies:

Number of actions (EC): The same holds true for follow-up information requested by ECs, i.e. 80 % (a) of all applications. However, in most Member States such requests are channelled, as the ‘single opinion’, via an EC. Therefore, the number of applications equals the number of NCAs, i.e. 9 763 (b). The number of cases handled is therefore: a*b = 7 810.

1.13. Implementation costs

In terms of resources of the Agency, the impact would be as follows:

A central assessment would apply to all clinical trials planned in the EU, whether mono-national or multinational. This scope is necessary to ensure that the main benefit of this policy option materialises, i.e. easier roll-out of a clinical trial in an additional Member State.

The assessment would not be carried out by Agency staff, but by rapporteurs from Member States. However, Agency staff would coordinate this process. This compares with the centralised marketing authorisation. Every year approximately 100 applications (a) are submitted to the Agency. Every year, approx. 850 major changes (Line extension or major variation Type II) (b) to the marketing authorisation application dossier are submitted subsequently to the granting of the marketing authorisatoin of the Commission. The Agency has approx. 50 FTEs (c) to coordinate the initial authorisation process, and 65 FTEs (d) for subsequent changes to the variations.

On the basis of these figures, and considering the number of initial applications (approx. 4 400) (e) and follow-up changes (SAs, approx. 24 000) (f) to clinical trials, the staff need wold be as follows:

Initial application: c*e/a = 2 200 FTEs (g)

Subsequent changes: f*d/b = 1 835 FTEs (h)

Total: g + h = 4 035 FTEs

              Annex 4: Objective No 2 — Regulatory requirements adapted to practical considerations and needs

1.           Policy option No 2/1 — calculation of baseline

Obligatory insurance

Administrative costs

Collecting information regarding the current rules, putting papers and documents together, filling in forms, sending them, etc. in order to comply with national insurance/indemnity requirements takes, on average, approximately 4 man-hours per application to conduct a clinical trial (a).[77] One man-hour costs 60 EUR (b).

In view of the 7 963 applications per year (2010) (c), the administrative costs for insurance/indemnity requirements are:

a*b*c = 1 911 120 EUR.

Other compliance costs

Example: Costs per patient per year for insurance in different Member States (in EUR):[78]

Belgium || 14.50

France || 75.00

Germany || 75.00

Italy || 50.00

The Netherlands || 23.00

On the basis of these figures, along with other figures submitted in the 2011 public consultation,[79] it can be assumed that the average costs of insurance per participant in a clinical trial are 50 EUR per year (a).

As a clinical trial lasts, on average, approximately 3 years, in view of the number of subjects planned for recruitment (see Annex 2), it can be deduced that at any given time approximately 1 500 000 patients are enrolled in clinical trials (b).

Consequently, the other compliance costs per year are:

a*b = 75 m EUR.

Number of claims/level of damages

There are very limited figures on the number of damages claims. In any case, damages claims are extremely rare.

– Data from one insurance company in the Netherlands show that over a period of nine years 14 claims were granted. The total amount of compensation for these cases was 43 000 EUR. The administrative costs for the insurance company totalled approximately 38 000 EUR. The total costs for the policy are approximately 235 000 EUR.[80]

In the Netherlands, every year between 350 and 650 clinical trials are applied for (2006: 445; 2007: 638; 2008: 642; 2009: 358). According to EudraCT, since the Clinical Trials Directive came into force, enrolment of 232 661 participants in clinical trials has been planned.

– The German ‘KKS Netzwerk — Koordinierungszentren für klinische Studien’ reported, over a period of 10 years (1997-2007) involving more than 20 000 trial subjects, three liability cases with minor damages.[81]

– The ‘Insurance Working Group’ of the Permanent Working Party of Research Ethics Committees in Germany reported that every year about 80 to 100 new liability claims are investigated. Between 2005 and 2010, recruitment of 700 000 subjects was planned, i.e. approximately 117 000 per year. In most of the cases where liability was accepted the sum was low, but in a very few an amount of more than 100 000 EUR has been paid in compensation in recent years.[82]

– Between 2005 and 2010, the Finnish Patient Insurance Centre and the Finnish Pharmaceutical Insurance Pool handled 19 claims for compensation, of which four led to compensation payments.[83] According to EudraCT, between 2005 and 2010 enrolment of 299 059 participants in clinical trials was planned in Finland, i.e. 50 000 per year.

– According to the Danish Patient Insurance System (DPIS), over a period of 10 years out of 49 claims for compensation by patients participating in clinical research projects 27 were accepted. This added up to a total of approximately 550 000 EUR.[84] According to EudraCT, from the entry into force of the Clinical Trials Directive until 2010 enrolment of approximately 120 000 participants in clinical trials was planned in Denmark, i.e. approximately 20 000 per year.

– The European Organisation for Research and Treatment of Cancer (EORTC) reported that, in the five years up to 2011, ten damages claims from two countries (only one of which was a Member State) led to indemnity of 60 000 EUR. This was for a population of approximately 30 000 subjects recruited in 43 clinical trials.[85]

In view of the foregoing, the following assumptions can be made:

Damages claims

The figures from Denmark, Germany and Finland show that between 0.006 % and 0.08 % of subjects (DK: 0.0245 %, DE: 0.08 %, FI: 0.00635 %) claim damages.

For the purposes of this impact assessment, it will be estimated that, as an EU average, 0.05 % of recruited subjects claim damages — be it successfully or not.

Compensation granted

The question of whether, following a claim, compensation is actually paid depends strongly on the civil law systems in the Member States. On the basis of the figures set out above, it can be assumed that approximately 50 % of the claims lead to compensation being granted, i.e. to 0.025 % of all subjects enrolled.

Level of compensation

The figures set out above show that, on average, damages claims range from 3 000 to 6 000 EUR. This assumption is in line with various estimates made in publications[86] and discussions in conferences.[87]

Annual safety report

Administrative costs

The administrative costs for the annual safety report are indicated in Annex 3 (2 x 2 636 010 EUR = 5 272 020 EUR).

Other compliance costs

Apart from the administrative costs, there are the costs of the actual drafting and setting-up of the report. The ASR requires approximately 40 man-days, i.e. 320 man-hours (a). This already factors in the fact that, over the years, the efforts for drafting decrease: in subsequent years, only an update of the report for the previous year is required. One man-hour costs 60 EUR (b).

The report to be submitted is largely identical in format and content. Moreover, practically all Member States accept the report in English, i.e. without any need for a translation.

The report addresses subject safety in the light of the investigational medicinal product. It can be assumed that each year approximately 7 700 ASRs are drafted (see Annex 2, point 8) (c).

The other compliance costs per year for the annual safety report are therefore:

a*b*c = 147.8 m EUR.

2.           Policy option No 2/2 — Enlarging the scope of non-interventional studies

In 2010, a total of 707 Phase IV clinical trials were authorised (a). These involved 1 029 applications (b) and 52 230 patients (c).[88]

If the definition of ‘non-interventional study’ were enlarged, it can reasonably be assumed that approximately 50 % of these phase IV trials, the associated applications and the patients participating would be freed of the obligations imposed by the Clinical Trials Directive.

Obligatory insurance

Administrative costs

Collecting information regarding the current rules, putting papers and documents together, filling in forms, sending them, etc. in order to comply with the submission requirements for national insurance and indemnity takes approximately 4 man-hours (d). One man-hour costs 60 EUR (e).

On this basis, this means the following savings in administrative costs:

b/2*d*e = 123 480 EUR.

Other compliance costs

As indicated above, the average cost of insurance per participant is 50 EUR per year (g).

The 707 phase IV trials approved in 2010 involved 52 230 patients (c). As indicated above, participation in a clinical trial lasts, on average, approximately 3 years (i).

The yearly savings in other compliance costs are therefore:

c/2*i*g = 3.92 m EUR.

Annual safety report

Administrative costs

Collecting information regarding the current rules, putting papers and documents together, filling in forms, sending them, etc. in order to comply with the requirements for the annual safety report takes approximately 1.5 man-hours (k). One man-hour costs 60 EUR (l).[89]

On average, each clinical trial is conducted over a period of approximately 3 years (m).

This means that this policy option would bring about the following savings in administrative costs:

a/2*m*k*l = 95 445 EUR.

Other compliance costs

As shown in Annex 2 (point 8), the approximately 12 000 ongoing clinical trials lead to the drafting of approximately 7 700 ASRs per year.

Of these ongoing clinical trials 3*707 (3 years is the average duration of a clinical trial)  are assumed to be phase IV trials, i.e. 2 121 clinical trials. 50% of these phase IV trials are of interest here, i.e. 1 060 clinical trials.

It follows that the amount of ASR which would not have to be drafted is

7 700*1 060/12 000 = 680 = (o)

As indicated above, drafting the report takes approximately 40 man-days, i.e. 320 man-hours (p). This already factors in the fact that, over the years, the efforts for drafting decrease: in subsequent years, only an update of the report for the previous year is required. One man-hour costs 60 EUR (q).

This means that this policy option would yield the following savings:

o*p*q = 13.06 m EUR.

3.           Policy option No 2/3 — Excluding ‘academic sponsors’

In 2010, some 1 620 clinical trials by ‘academic sponsors’ were authorised (a).[90] These involved 2 037 applications (b) and 93 242 patients (c).[91]

Obligatory insurance

Administrative costs

As indicated above, collecting information regarding the current rules, putting papers and documents together, filling in forms, sending them, etc. in order to comply with national insurance/indemnity requirements takes, on average, approximately 4 man-hours per application (d).[92] One man-hour costs 60 EUR (e).

In view of the 2 037 applications per year (2010) (b), the savings in administrative costs under this policy option, compared with the baseline, are:

b*d*e= 488 880 EUR.

Other compliance costs

As indicated above, the average cost of insurance per participant in a clinical trial is 50 EUR per year (g).

The 1 620 trials approved in 2010 involved 93 242 patients (c). As indicated above, participation in a clinical trial lasts, on average, approximately 3 years (i).

The yearly savings in other compliance costs are therefore:

c*i*g = 13.99 m EUR.

Annual safety report

Administrative costs

Collecting information regarding the current rules, putting papers and documents together, filling in forms, sending them, etc. in order to comply with the requirements for the annual safety report takes approximately 1.5 man-hours (k). One man-hour costs 60 EUR (l).[93]

On average, each clinical trial is conducted over a period of approximately 3 years (m).

This means that this policy option would yield the following savings in administrative costs:

a*k*l*m = 437 400 EUR.

Other compliance costs

As shown in Annex 2, the approximately 12 000 ongoing clinical trials lead to the drafting of approximately 7 700 ASRs per year.

Of these ongoing clinical trials 3*1620 are assumed to be sponsored by 'academic sponsors', i.e. 4 860 clinical trials.

It follows that the amount of ASR which would not have to be drafted is

7 700*4 860/12 000 = 3118 = (o)

As indicated above, drafting the report takes approximately 40 man-days, i.e. 320 man-hours (p). This already factors in the fact that, over the years, the efforts for drafting decrease: in subsequent years, only an update of the report for the previous year is required. One man-hour costs 60 EUR (q).

This means that this policy option would yield the following savings:

o*p*q = 59.9 m EUR.

4.           Policy option No 2/4 — Removing regulatory requirements on the basis of the knowledge of the IMP

Phase IV clinical trials are, by definition, clinical trials with authorised medicines. However, not each and every phase IV clinical trial is limited to the authorised indication. Moreover, certain phase I to III clinical trials may be performed with medicines whose active ingredient is already contained in authorised medicines.[94]

For the purposes of this assessment, it is therefore assumed that all Phase IV clinical trials, and only Phase IV clinical trials, involve authorised IMPs in the authorised indication.

In 2010, a total of 707 Phase IV clinical trials were authorised (a). These involved 1 029 applications (b) and 52 230 patients (c).[95]

Obligatory insurance

If the obligatory insurance were waived for clinical trials with authorised IMPs, the following would apply:

Administrative costs

Collecting information regarding the current rules, putting papers and documents together, filling in forms, sending them, etc. in order to comply with the national insurance/indemnity requirements takes approximately 4 man-hours (d). One man-hour costs 60 EUR (e).

On the basis of the foregoing, this means the following annual savings in administrative costs:

b*d*e = 246 960 EUR.

Other compliance costs

As indicated above, the average cost of insurance per participant in a clinical trial is 50 EUR per year (g).

The 707 phase IV trials approved in 2010 involved 52 230 patients (c). As indicated above, participation in a clinical trial lasts, on average, approximately 3 years (i).

The yearly savings in other compliance costs are therefore:

c*i*g = 7.84 m EUR.

Annual safety report

Administrative costs

Collecting information regarding the current rules, putting papers and documents together, filling in forms, sending them, etc. in order to comply with the requirements for the annual safety report takes approximately 1.5 man-hours (k). One man-hour costs 60 EUR (l).[96]

On average, each clinical trial is conducted over a period of approximately 3 years (m).

This means that this policy option would yield the following savings in administrative costs:

a*l*b*k = 190 890 EUR.

Other compliance costs

As shown in Annex 2, the approximately 12 000 ongoing clinical trials lead to the drafting of approximately 7 700 ASRs per year.

Of these ongoing clinical trials 3*707 are assumed to be sponsored by 'academic sponsors', i.e. 2 121 clinical trials.

It follows that the amount of ASR which would not have to be drafted is

7 700*2 121/12 000 = 1360 = (n)

As indicated above, drafting the report takes approximately 40 man-days, i.e. 320 man-hours (o). This already factors in the fact that, over the years, the efforts for drafting decrease: in subsequent years, only an update of the report for the previous year is required. One man-hour costs 60 EUR (p).

This means that this policy option would yield the following savings:

n*o*p = 26.1 m EUR.

5.           Policy option No 2/5 —Insurance/Optional ‘national indemnification mechanism’

This policy option provides for an optional national indemnification mechanism. The calculation of the costs and risks related to this mechanism is based on the assumption that all sponsors make use of this mechanism: indeed, in view of how difficult it is for sponsors to obtain insurance cover, it is very likely that they will practically all opt in to this mechanism.

Administrative costs

The national indemnification mechanism would make the difficult researching for the national requirement superfluous. Instead, opting in to the mechanism would suffice to prove that the patient is covered for damages. A simple document could prove this. Obtaining and submitting this document would generate administrative costs of 0.5 man-hours (a) per application. One man-hour costs 60 EUR (b).

In view of the 7 924 applications per year (2010) (c), the administrative costs for insurance/indemnity requirements are:

a*b*c = 237 720 EUR.

Implementation costs for the national indemnification mechanism

Running costs to maintain the national indemnification mechanism

Some Nordic Member States have set up a compensation scheme like that proposed in this policy option. Their experience teaches lessons about the costs for running such a scheme — apart from the actual costs to cover damages.

These costs must be assessed on the basis of the claims, not of the cases where damages were granted. For example, in Denmark, the Patient’s Insurance Association deals with approximately 8 000 claims (d) for damages per year (clinical-trial-related and other). To deal with all incoming claims, the Association has approximately 130 staff (e).

As shown above, approximately 0.05 % of all enrolled subjects claim damages in a given year (f). Each year, 400 000 subjects (g) are enrolled in clinical trials in the EU. A clinical trial is assumed to last, on average, 3 years (h).

The total claims per year in the EU are therefore:

f*g*h = 600 = i.

These figures allow some extrapolation and show that the running costs, in terms of staff, of a national indemnification mechanism in the EU per year are:

e*i/d= 9.75 FTEs.

This means costs of approximately 9.75*70 000 EUR per year, i.e. 682 500 EUR (o).

It is assumed that the costs for this personnel are recouped by Member States via fees.

Costs for covering damages

The national indemnification mechanism, i.e. Member States, would have to bear the costs for damages occurring in clinical trials in the Union.

Every year 400 000 clinical trial subjects can be expected to participate in a trial (2010) (k). Each clinical trial lasts approximately 3 years (l).

As shown above, it can be assumed that 0.025 % of participants justifiably claim damages linked to a clinical trial (f).

On average, damages claims range from 3 000 to 6 000 EUR. For the purposes of this calculation, a value of 4 500 EUR will be assumed (n).

This means that the costs of damages per year are:

k*l*f*n = 135 000 EUR = (p).

It would be left to each Member State to decide whether and how it intends to cover these costs.

Other compliance costs

It is assumed here that implementation costs are borne, by way of fees, by the sponsors. These costs would be, per year:

o+p = 817 500 EUR.

6.           Policy option No 2/6 —Combination of policy option No 2/4 and No 2/5

Regarding annual safety report:

As policy option No 2/5 does not have an impact on the obligation to draft and submit an annual safety report, the economic impact in comparison to the baseline is identical as under policy option No 2/4.

Regarding obligatory insurance/indemnification:

Regarding administrative costs, according to policy option No 2/4, 1029 clinical trials applications (a) would not be covered by the obligatory insurance/indemnification: As set out under policy option No 2/5, the administrative costs in a national indemnification mechanism would be 0.5 man-hour (b) per application with a value of 60 EUR per hour (c).

This means that, in terms of administrative costs the additional savings compared to policy option No 2/5 are

a*b*c = 30 870 EUR

The impact of this policy option on other compliance costs compared to the baseline is identical to policy option No 2/5. Wee the impact assessment report for more explanation.

              Annex 5: Objective No 3 — Addressing the global dimension of clinical trials when ensuring compliance with GCP

1.           Policy option No 3/2: Facilitating GCP inspections by increasing transparency (database of all clinical trials)

The costs of this policy option are limited to administrative costs.

It is estimated that approximately 30 % (a) of all clinical trials requested in an application for an EU marketing authorisation are conducted exclusively in non-EU countries. Each year, about 100 applications (b) are submitted to the Agency,[97] each of which refers, on average, to approximately 100 clinical trials (c).

In addition, every year approximately 2 000 (d) applications for national marketing authorisation are submitted.[98] Each involve, on average, 10 clinical trials (e). Of these, it can be estimated that 20 % are performed exclusively in non-EU countries (h).

Publication of this information in an official public register takes approximately 2 man-days, i.e. 16 man-hours (f). One man-hour costs approximately 60 EUR (g).

Consequently, the administrative costs for this policy option are:

a*b*c*f*g + d*e*h*g*f = 2.88 m EUR + 3.84 m EUR = 6.72 m EUR.

2.           Policy option No 3/3: Inspections of the third countries' regulatory systems for clinical trials

Currently there are no inspection capacities at EU level foreseen. However, a somewhat comparable capacity exists at EU level for system inspections (audits) in the food and veterinary sector: the Food and Veterinary Office of the European Commission ('FVO'). In 2010 the FVO performed 248 audits, of which 105 were in non-EU countries. The inspections in non-EU countries cost approximately 800 000 EUR. The other compliance costs, including costs for staff, must be added to this: In terms of staff, the 248 FVO audits in 2010 were performed by 85 auditors, backed up by an additional 52 support staff.

On the basis of these figures one can extrapolate that, for approximately 8 system inspections per year, the following resources would be required:

· Staff: 3 FTE (inspectors), plus 2 FTE (support staff);

· Costs for conducting inspections: approximately 76 000 EUR.

To finance this policy option, the following approaches shall be discussed here:

· Financing through fees is not conceivable, as system inspections would require fees to be paid by the government of a third country.

· Subsidies through the EU budget (structural) is difficult to envisage in view of the political commitment of the Commission to reduce staffing level at EU institutions.

· Cross-subsidies from fees for assessment of the marketing authorisation application. This financing strategy would only be possible if the system inspection was located with the Agency. In technical terms, the legislative framework for fees would have to be amended. In political terms, this approach would lead to an unfair distribution of burden for those actors who pursue (obligatorily or voluntarily) the centralised marketing authorisatoin procedure. Moreover, it would not correspond to the principle of "fee for service".

· Re-allocation of existing resources within the Commission or the Agency: Within the Commission, a re-allocation of resources on the scale set out here is enviseagable.

3.           Policy option No 3/4: GCP inspections of non-EU countries' clinical trial sites by the Agency

As for policy option No 3/3, currently there are no inspection capacities dedicated to GCP inspections by the Agency or the Commission.

Data from Member States show that a team of 6 GCP-inspectors (plus support staff) can perform approximately 55 inspections per year, including 12 inspections in third countries.[99] GCP inspections in non-EU countries require typically more preparatory time, as well as more travel time, than domestic inspections. Therefore, it is assumed that one inspector-FTE can conduct 6 GCP inspections in non-EU countries per year.[100]

The number of sites included in pivotal clinical trials submitted to the Agency in the context of marketing authorisations is approximately 8 000 per year.

As set out in the impact assessment report, it is assumed, that only 10% of these sites are inspected. An extrapolation of the figures above shows that this approach would require approx. 1 300 FTE in inspectors, plus support staff.

In view of the present difficulties to obtain additional resources at EU level, it is difficult to conceive an increase in the range set out in this policy option.

              Annex 6: Costs and financing strategies for the single submission point

This annex presents the details of the costs for the single submission point (1) and possible ways to finance it (2).

1. Costs for a single submission point

During the impact assessment process, the Agency has been consulted on the costs of a single submission point. Equally, Commission inhouse expertise has been sought.

As a result two different approaches could be pursued:

           'Extensive IT solution':

The Agency would pursue an 'extensive IT solution' including user validation functionalities, an IT helpdesk, a business support helpdesk, and operational support.

The one-off costs for an 'extensive IT solution' would be 6.3m EUR (a) (including 0.6m EUR (b) for updating the pharmacovigilance system for SUSAR reporting).

Running costs would be 20% (c) of the one-off costs per year. Thus running costs (excl. staff) would thus be

a*c = 1.26m EUR

To this adds, in terms of human resources, 11 Administrator posts (g) and 8 Assistant posts (d). These FTEs do not include the FTEs referred to in policy option No 1/3. According to previous calculations of the Agency, costs per FTE are 153 226 EUR (AD, e) and 81 617 EUR (AST, f). To this adds an overhead of 48.5%.

The running annual costs (incl. staff) of the 'extensive IT solution' would be thus

a*c + g*e*1.485 + d*f*1.485 = 4.73m EUR

           'Limited IT solution':

The Commission would pursue a 'limited IT solution' which would include less support activities, such as Helpdesks for IT. Moreover, it would build on existing IT functionalities within the Commission.

Only the update of the pharmacovigilance system would remain with the Agency, as the Agency has already a pharmacovigilance IT system in place.[101]

In this case, one-off costs would be 1.02m EUR, plus 0.6m EUR for pharmacovigilance, i.e. 1.62m EUR.

Running costs would be 0.22m EUR per year, plus 20% of 0.6m EUR for pharmacovigilance, i.e. 0.34m EUR per year. These costs include staff requirements for the programming.

In this 'limited IT solution', an additional 0.25 FTEs would be required to support the programming activity in terms of regulatory expertise.

2. Financing strategies

When looking at strategies as to financing these costs, there are three viable possibilites which shall be presented below. Of these three possibilities, two are only workable if the political decision was taken to allocate the single submission point within the Agency.

a. Cross-subsidy from fees for marketing authorisation activities of the Agency (only possible if single submission point is allocated with the Agency)

This approach would impose the costs for the single submission point on the pharmaceutical companies whose products have to be assessed by the EMA (rather than national agencies) prior to marketing authorisation by the Commission.

The EMA conducts approximately 100 assessments in connection with marketing authorisations per year. The fee for such assessments is currently approximately 260 000 EUR.

In view of the costs (see point 1; it is assumed that the one-off costs occur in the first three years), the authorisation fee would have to increase by 79 000 EUR, i.e. by approximately 30%. After the first three years, the fee would rise by 58 600 EUR per marketing authorisation application, i.e. by approximately 25%.

This approach would mean that a relatively small number of companies would bear the burden for a tool which is of benefit not only to them, but also to their competitors and academic researchers. Moreover, it would not correspond to the principle of "fee for service".

In technical terms, amendment of the ‘Fees Regulation’ (Regulation (EC) No 297/95) would be required.

b. Separate fee for all applicants for approval of a clinical trial (‘28th fee’ ‑ only possible if single submission point is allocated with the Agency)

This approach would entail a separate fee in addition to the national fees (potentially 27 national fees, plus the EU fee).

In view of the 4 400 clinical trials per year, if the fee was only imposed at the moment of the application for authorisation of a clinical trial that fee would have to be, in the first three years, 1 800 EUR. After the first three years, the fee would have to be 1 330 EUR.

The critical point would be the collection of fees which, in itself, requires an important amount of resources. These resources are largely independent of the sum collected by way of fees.

c. Support from the EU budget (possible no matter if single submission point is allocated with the Agency or with the Commission)

This approach would entail a subsidy from the EU budget to set up and maintain the single submission point.

Regarding the EU budget, the Commission has proposed the EU 'public health program' on 9 November 2011.[102],[103]  If the program is adopted as proposed by the Commission, it could potentially provide the financial means to finance the 'limited IT solution'. However, the financial means available through this program would not allow for financing the 'extensive IT solution'.

              Annex 7: Policy option 1/3: Support structure at EU level - implementation costs

1. Costs at EU level

Apart from the costs for the single submission portal (see Annex 6), the implementation costs would be linked to technical support and the role of a 'facilitator' of the joint assessment.

As with the single submission portal (see Annex 6) the Agency has been consulted on the costs of implementation at EU level. Also in this case there are two possible approaches:

'Extensive support structure': according to estimations of the Agency, its resource requirements would be 7 FTEs (3 administrators (a) and 4 assistants (b)). According to previous calculations of the Agency costs per FTE are 153 226 EUR for administrators (c) and 81 617 EUR per assitstant (d). To this adds an overhead of 48.5%.

The running annual staff costs of the large-scale solution of the Agency would be thus:

a*c*1 485 + b*d*1 485 = 1.17 m EUR

'Limited support structure': the Commission would pursue a limited support structure. Such structure would require, in addition to the existing available resources of 0.25 FTE (see above) an additional resource of 1.50 FTEs (all administrators, including overhead).

Additional costs for travel expenses: It is assumed that one meeting every two months is necessary to deal with all structural and cross-cutting issues. One delegate per Member State would be reimbursed.[104]

If the support structure is provided by the Agency meetings would take place in London. The Agency would calculate 1 300 EUR per delegate per meeting (i.e. 6*27*1 300 EUR = 210 600 EUR per year).

If the support structure is provided by the Commission, the meetings would take place in Brussles. The Commission would calculate, in accordance with the applicable rules for the Commission, costs of 630 EUR per delegate per meeting (i.e. 6*27*630 EUR = 102 060 EUR).

2. Financing strategies

In view of the political commitment of the Commission to reduce staffing level at EU institutions,[105] an increase in staff can be pursued only as follows:

· Fees or cross-subsidies as set out in Annex 6 (2), points (a) and (b). These two financing strategies would only be possible if the support structure was located with the Agency.

· Re-allocation of existing human resources within the Commission or the Agency.

              Annex 8: Involvement of Commission or Agency – Key points for consideration

The impact assessment addresses additional tasks for EU-bodies in three contexts:

           Setting-up and maintaining a single submission point; and

           Technical support and 'facilitator' of the joint assessment, as referred to in point Error! Reference source not found. of the report;

           'Systems inspections', as referred to in point Error! Reference source not found. of the report.

In all three contexts the impact assessment leaves open whether these tasks should be performed by the Agency or the Commission. This decision is left to political decision-making on the basis of the aid and information provided in this impact assessment.[106]

In this respect the following arguments and counter-arguments in favour and against both the Agency and the Commission should be born in mind:

1. Setting up and maintaining the single submission point

           Costs and financing: The estimated costs are in a large range depending on the IT solution chosen (see Annex 6). It is not clear whether the Agency would be in a position to pursue a 'limited IT solution' in view of the costs estimated by EMA.

              Allocation of the single submission point with the Agency gives a broader range of means to finance this IT solution, such as fees (see Annex 6). On the other hand, as set out in Annex 6, it is not certain whether financing tools other than the EU budget are viable.

           Experience: The Agency is already today in charge of programming and maintaining EudraCT. EudraCT contains information on all clinical trials for which a request for authorisation has been submitted. Some information contained in EudraCT has been made public through the ClinicalTrialsRegister.eu.

              EudraCT could be used as a starting point for the single submission point.

              On the other hand, it is far from certain if the IT framework for EudraCT can support the functionalities required for a single submission point. In this respect the Agency has highlighted that a strategy of a completely new system may be more cost efficient in the longer term.

              Moreover, the Commission, and in particular the lead-service DG SANCO, has experience with similar systems of submission points in in other policy areas.

           Synergies with medical devices legislation: The revision of the medical devices legislation is ongoing. Currently, it is being considered to introduce, as regards clinical experiments with medical devices (so-called 'clinical investigations'), a single submission point, too. While a final decision as to where this point is allocated is still to be made (the impact assessment on the recast of the medical devices Directive has left this open)[107], it would be preferable that both submission points are allocated with the same body (Commission or Agency). This would create (cost-saving) synergies. It would also facilitate a coherent message to stakeholders.

2. Technical support and 'facilitator' of the joint assessment

Both the Commission and the Agency have experience in this type of activity. In particular, the existing fora (see point Error! Reference source not found.) can be considered as equivalent to the technical support and 'facilitator' provided in this policy option.

In terms of costs, the difference would be limited to higher travel expenses costs if the support function would be allocated with the Agency (see Annex 7).

3. 'System inspections'

While the Agency has strong experience in the coordination of GCP inspections, the Commission has experience with the conduct of 'system-inspections' in non-EU countries - albeit in a different area (food and veterinary sector).

In terms of costs, the Commission would be in a position to re-allocate to this task some resources currently located in the Commission.[108]

[1]               Report of Working Group VI of the Council for International Organisations of Medical Sciences (CIOMS), ‘Management of safety information from clinical trials’, 2005, p. 232. A more detailed overview, with examples, is contained in ICH Guideline E8, ‘Note for guidance on general considerations for clinical trials’ (CPMP/ICH/291/95).

[2]               Cf. recital 10 of the Clinical Trials Directive.

[3]               Commission Directive 2005/28/EC of 8 April 2005 laying down principles and detailed guidelines for good clinical practice as regards investigational medicinal products for human use, as well as the requirements for authorisation of the manufacturing or importation of such products (OJ L 91, 9.4.2005, p. 13).

[4]               http://ec.europa.eu/enterprise/pharmaceuticals/eudralex/vol10_en.htm.

[5]               Article 2(i) of the Clinical Trials Directive refers to participants in clinical trials as ‘subjects’. In this document, the term ‘participants’ is used.

[6]               OJ L 311, 28.11.2001, p. 67.

[7]               Also referred to as ‘non-interventional studies’ or ‘observational studies’.

[8]               Response of the European Cancer Patients Coalition (ECPC) to the 2009/10 public consultation, page 5.

[9]               EFPIA statement at: http://www.efpia.eu/content/default.asp?PageID=507.

[10]             Response of the European Confederation of Pharmaceutical Entrepreneurs (EUCOPE) to the 2009/10 public consultation, p. 4.

[11]             Response of the European biopharmaceutical enterprises asociation (EBE) to the 2009/10 public consultation, p. 1.

[12]             Response of the Federation of the European Academies of Medicines (FEAM) to the 2009/10 public consultation, page. 1.

[13]             Press release of the ESF and the EMRC, 30 January 2012; http://www.esf.org/media-centre/ext-single-news/article/improving-medical-treatment-requires-a-risk-based-approach-to-the-regulation-of-clinical-trials-799.html.

[14]             Cf. the literature review in ICREL (pp. 25-43). However, there are also publications which take a less black-and-white approach when discussing the negative impact of the Clinical Trials Directive (cf. Berendt et al., ‘Effect of the European Clinical Trials Directive on academic drug trials in Denmark: retrospective study of applications to the Danish Medicines Agency 1993-2006’, BMJ, published online on 6 December 2007).

[15]             Clin Trials, 20 August 2010, p. 1.

[16]             BMJ 2006; 332 doi: 10.1136/bmj.332.7540.501 (Published 2 March 2006)

[17]             Morice AH, Lancet 361:1568 (2003).

[18] http://www.hma.eu/fileadmin/dateien/HMA_joint/HMA_Strategy_Paper_II/HMA_Strategy_final_version__2_.pdf.

[19]             'Statement’ of the governments of the Netherlands, Belgium, Bulgaria, Ireland, Spain, Finland and Sweden, annexed to the EU pharmacovigilance legislation adopted in 2010, Document 10779/10, 22 June 2010 ADD1 (http://register.consilium.europa.eu/pdf/en/10/st10/st10779-ad01.en10.pdf).

[20]             P. 1. See also the open letter, in July 2008, from the UK Secretary of State for Business, Enterprise and Regulatory Reform on ‘25 ideas for simplifying EU law’: http://www.administrative-burdens.com/filesystem/2008/07/25_ideas_for_simplifying_eu_law_517.pdf.

[21]             Page 9 (http://www.consilium.europa.eu/uedocs/cms_data/docs/pressdata/en/lsa/118278.pdf).

[22]             http://www.europarl.europa.eu/sides/getDoc.do?type=TA&reference=P6-TA-2008-0121&language=EN.

[23]             Le Soir, Belgium, 4 Septembre 2010, p. 8: 'Les essais cliniques belges en crise'.

[24]             The Daily Telegraph, 13 January 2012, p. 20.

[25]             The Telegraph, 26 September 2011.

[26]             Tröhler, Ulrich; The long road of moral concern: Doctors' Ethos and Statue Law relating to Human Research in Europe; History and theory of Human Experimentation (Eds. Schmidt, Ulf; Frewer, Andreas), 2007, p. 36.

[27]             http://www.wma.net/e/.

[28]             http://ec.europa.eu/enterprise/pharmaceuticals/eudralex/vol-10/3cc1aen.pdf.

[29]             http://www.emea.europa.eu/pdfs/human/ich/013795en.pdf.

[30]             See, for example, http://www.emea.europa.eu/htms/human/ich/ichefficacy.htm.

[31]             http://conventions.coe.int/treaty/Commun/ChercheSig.asp?NT=005&CM=&DF=&CL=ENG.

[32]             http://conventions.coe.int/treaty/en/treaties/html/164.htm.

[33]             It should be noted that EudraCT data is supplied by the NCAs of MSs. The data submitted prior to 2007 has some flaws. This is due to the fact that – despite the transposition date of 1 May 2004 – the Clinical Trials Directive (and the corresponding reporting requirements to EudraCT) was fully applied only in 2006 or later.

[34]             EMA/INS/GCP/154352/2010, 5 November 2010.

[35]             The sponsor status 'commercial' or 'non-commerical' is self-declared by the sponsor. In the absence of a EU-definition of these terms, usually a formal criteria (company or not) applies.

[36]             These numbers do not match the total of the entries in table 1. The reason is that the (self-declared) phase is sometimes missing. The data for 2011 is a forecast on the basis of EudraCT data available on 18 October 2011.

[37]             The same clinical trial can be applied for in several MS, i.e. one clinical trial can mean up to 27 clinical trial applications.

[38]             The total per year is not absolutely identical with the figures in table 3. This can be explained by the fact that data in EudraCT is self-declared by the sponsors and as such not always fully stable.

[39]             For the terminology, see point Error! Reference source not found. of the impact assessment report. On the detailed terminology see ICH Topic E8 – General Considerations for Clinical Trials (CPMP/ICH/291/95).

[40]             The total per year is not absolutely identical with the figures in table 6. This can be explained by the fact that data in EudraCT is self-declared by the sponsors and as such not always fully stable.

[41]             Commission survey amongst NCAs. Substantial amendments (SAs) submitted: the SAs received by the NCA. This excludes by definition (see COM guideline CT-1) SAs submitted for assessment by Ethics Committees, and 'SAs for information only'

[42]             AT: Classification of the amendment is the remit of the sponsor. This number therefore includes all submission classified by the sponsor as “substantial”.

[43]             According to the Italian legislation, the NCA is responsible for the authorization of phase I (ISS) and Advanced Therapy Investigational Medicinal Products clinical trials (AIFA) substantial amendments. When the NCA is not involved, the coordinating Ethics Committee is responsible for the assessment of the amendments at a national level. When considering all substantial amendments assessed by the NCA or the coordinating EC, the number would be 2,086. 

[44]             ICREL, p. 74.

[45]             Getz, Zuckerman, Cropp, Hinle, Krauss, Kaitin, ‘Measuring the incidence, causes and repercussions of protocol amendments’, Drug Information Journal, Vol. 45, p. 265 (2.3 amendments per protocol amended).

[46]             Commission survey amongst NCAs. SUSAR: A given suspected unexpected serious adverse reaction. SUSAR reports: the reports received by the NCA. One SUSAR may trigger many reports (for example, follow-up reports or double reporting from different reporting stakeholders).

[47]             AT: This is an estimation of the total number of received reports. It is not possible to extract the number of unique reports from the Austrian database.

[48]             AT: National reporting.

[49]             AT: Direct reporting to Eudravigilance.

[50]              Data generated from the EudraVigilance Datawarehouse as Belgium has no national database.

[51]             Data refers to national SUSARs and are generated from the EudraVigilance Datawarehouse as Italy has no national database.

[52]             According to implementing guidance of the Commissoin, the ASR is based on the IMP.

[53]             Survey of the Commission amongst NCAs. Annual Safety Reports (ASRs): These are the number of 'unique reports'. The number of 'received reports' may be higher: In practice a sponsor may submit the identical 'unique report' to the same NCA several times.

[54]             AT: This is an estimation of the total number of received reports. It is not possible to extract the number of unique reports from the Austrian database.

[55]             This is an estimation of the total number of reports. It is not possible to extract the number of unique reports from the Belgian database.

[56]             This is an estimation of the total number of reports.

[57]             This is an estimation of the total number of reports.

[58]             Practically all Member State authorities accept the ASR submitted in the English language.

[59]             ‘The pharmaceutical industry in figures’ (2010), European Federation of Pharmaceutical Industries and Associations.

[60]             For example, the purchase of the comparator can be a very important cost factor depending on the purchased medicinal product.

[61]             See point Error! Reference source not found. of the impact assessment report.

[62]             Costs are very valuable, from low to very high, and depend inter alia on therapeutic area.

[63]             OJ L 124, 20.5.2003, p. 36.

[64]             See Commission impact assessment report for the proposal of Directive 2011/62/EU (SEC(2008)2674, 10 December 2008), p. 74.

[65]             Article 2(3) of the Annex to Commissoin Recommendation 2003/361/EC.

[66]             Survey of the Commission amongst NCAs. If assessors are used which are actually attributed to other departments (for example, marketing authorization department), the 'share' of resources used for the purpose of assessing clinical trials should be indicated. Management staff (Head of unit etc.) should not be included.

[67]             European Forum for Good Clinical Practice Ethics Working Party (2007) Subgroup on Ethics Committees reviewing investigatoinal medicinal products with the European Union: the procedure for the ethical review of protocols for clinical research projects in the European Union (Int J Pharm Med 21:1-113 update 2008)

[68]             Submission by EUCROF in the 2011 public consultation.

[69]             According to a submission by Roche Pharmaceuticals in the 2011 public consultation, 101 hours for a clinical trial in one Member State and 159 hours for a clinical trial in three Member States.

[70]             Submission by the EUCROF in the 2011 public consultation.

[71]             Submission by EUCROF in the 2011 public consultation.

[72]             According to a submission by Roche Pharmaceuticals in the 2011 public consultation, 27 hours for a clinical trial in one Member State and 49 hours for a clinical trial in three Member States.

[73]             According to a submission by Roche Pharmaceuticals in the 2011 public consultation, 16 hours for a clinical trial in one Member State and 38 hours for a clinical trial in three Member States.

[74]             To the extent that this communication is a legal obligation.

[75]             Publications on the share of the costs of these specific aspects of clinical trials are limited. One public source is the report 'Clinical Trials in Poland' PriceWaterhouseCoopers, November 2010, p. 3 (http://www.pwc.com/gx/en/pharma-life-sciences/publications/clinical-trials-in-poland-2010.jhtml).

[76]             See Annex 2.

[77]             Submission by EUCROF in the 2011 public consultation.

[78]             Source: Submission by stakeholder.

[79]             See submission by the EORTC in the 2011 public consultation: depending on the Member State, costs range from 32 EUR to 250 EUR per person per year.

[80]             Source: Submission following workshop on clinical trials with the European Science Foundation.

[81]             Source: Submission following workshop on clinical trials with the European Science Foundation.

[82]             Submission by the Permanent Working Party of Research Ethics Committees in Germany in the 2011 public consultation.

[83]             Source: Ad hoc group on clinical trials.

[84]             Source: Submission following workshop on clinical trials with the European Science Foundation.

[85]             Submission by the EORTC in the 2011 public consultation.

[86]             Jungk, ‘Schadenersatzansprüche von Patienten in klinischen Prüfungen — ein Überblick’, DZKF, 7/8‑2007, p. 49.

[87]             See point Error! Reference source not found..

[88]             EudraCT.

[89]             This is calculated on the basis of a single submission point (cf. policy options No 1/2 to No 1/4).

[90]             EudraCT.

[91]             EudraCT.

[92]             Submission by EUCROF in the 2011 public consultation.

[93]             This is calculated on the basis of a single submission point (cf. policy options No 1/2 to No 1/4).

[94]             For example, bioequivalence studies.

[95]             EudraCT.

[96]             This is calculated on the basis of a single submission point (cf. policy options No 1/2 to No 1/4).

[97]             See also the Final Report of the European Medicines Agency (2010), p. 174 (http://ec.europa.eu/health/files/pharmacos/news/emea_final_report_vfrev2.pdf).

[98]             Idem, p. 53.

[99]             The efficiency of inspections of national clinical trial sites is higher than system-inspections in third countries.

[100]            This figure takes account of the fact that GCP inspections are usually conducted in a team.

[101]            Transferring this system to the Commission, as regards clinical trials, would create considerable inefficiencies.

[102]            COM(2011) 709.

[103]            Another possibility for financing the single submission portal that could be explored is via the EU program 'interoperability solutions for European public administrations' ('ISA' - Decision  No 922/2009/EC).

[104]            If, in a given Member States more than one national body is involved (for example, NCA and EC), that Member State has to find internal arrangements to ensure appropriate representation of views.

[105]            See Commission Communication 'A budget for Europe 2020', COM(2011)500, 29.6.2011, point 6.1.5. ('5% reduction in the staffing levels of each institution/service, agency and other body').

[106]            See point 1 of the European Commission impact assessment guidelines ('Impact assessment is an aid to political decision-making, not a substutitue for it').

[107]            Commission Staff Working Paper: Impact assessment – Revision of the regulatory framework for medical devices (not yet published), point 5.8.4. (Comparison of policy options 7A-7D).

[108]            For details, see Annex 5.

|| EUROPEAN COMMISSION ||

Brussels, 17.7.2012

SWD(2012) 200 final

VOLUME I

 

COMMISSION STAFF WORKING DOCUMENT

Impact assessment report on the revision of the “Clinical Trials Directive” 2001/20/EC

Accompanying the document

Proposal for a Regulation of the European Parliament and of the Council

on clinical trials on medicinal products for human use, and repealing Directive 2001/20/EC

{SWD(2012) 201 final}

{COM(2012) 369 final}

COMMISSION STAFF WORKING DOCUMENT

Impact assessment report on the revision of the “Clinical Trials Directive” 2001/20/EC

Accompanying the document

Proposal for a Regulation of the European Parliament and of the Council

on clinical trials on medicinal products for human use, and repealing Directive 2001/20/EC

Contents

1.     Procedural issues. 7

1.1.      Consultations of other Commission departments and agencies. 7

1.2.      Consultations of Member States. 7

1.3.      Stakeholder consultations. 7

1.4.      Contacts with non-EU authorities. 9

1.5.      Impact Assessment Board. 9

2.     Problem definition. 10

2.1.      Introduction — setting the scene. 10

2.1.1.       What are clinical trials?. 10

2.1.2.       The regulation of clinical trials in the EU.. 13

2.1.3.       Affected bodies and enterprises – the "sponsor" 14

2.1.4.       The benefits of clinical trials – Public and patient health. 14

2.2.      Problem identification. 17

Introduction. 17

2.2.1.       Separate submission, diverging assessments and regulatory follow-up of applications for clinical trials  21

2.2.2.       Greater difficulties with conducting clinical trials due to regulatory requirements not adapted to practical considerations and needs. 24

2.2.3.       Reliability of clinical trial data in a globalised research environment 26

2.3.      Union powers and subsidiarity. 28

3.     Objectives. 29

3.1.      Objective No 1: A modern regulatory framework for submission, assessment and regulatory follow-up  29

3.2.      Objective No 2: Regulatory requirements which are adapted to practical considerations, constraints and needs, without compromising the safety, well-being and rights of participants in clinical trials and without compromising data robustness. 30

3.3.      Objective No 3: Addressing the global dimension of clinical trials when ensuring compliance with GCP  30

3.4.      Coherence with strategic policy objectives of the EU and the Commission. 30

4.     Policy options. 31

4.1.      Objective No 1 — A modern regulatory framework for submission, assessment and regulatory follow-up of applications for clinical trials. 31

4.1.1.       Policy option No 1/1 — No action at Union level and reliance on voluntary cooperation of Member States (baseline option) 31

4.1.2.       Policy option No 1/2 — Single submission with separate assessment 32

4.1.3.       Policy option No 1/3 — Single submission with joint assessment by Member States of issues not related to ethical aspects. 32

4.1.4.       Policy option No 1/4 — Single submission with central assessment by the Agency of issues not related to ethical aspects. 34

4.1.5.       Policy option No 1/5 — Choice of legal form — Adopting the text of the Clinical Trials Directive in the form of a Regulation. 35

4.1.6.       Policy option No 1/6 — Combination of policy option No 1/3 and No 1/5. 35

4.2.      Objective No 2 — Regulatory requirements adapted to practical considerations and needs. 35

4.2.1.       Policy option No 2/1 — No action at Union level (baseline option) 35

4.2.2.       Policy option No 2/2 — Enlarging the scope of non-interventional trials. 35

4.2.3.       Policy option No 2/3 — Excluding ‘non-commercial sponsors’ 36

4.2.4.       Policy option No 2/4 — Removing regulatory requirements on the basis of the knowledge of the IMP  36

4.2.5.       Policy option No 2/5 — Insurance/Optional ‘national indemnification mechanism’ 37

4.2.6.       Policy option No 2/6 — Combination of policy option No 2/4 and No 2/5. 38

4.3.      Objective No 3 — Addressing the global dimension of clinical trials when ensuring compliance with GCP  38

4.3.1.       Policy option No 3/1: Leaving the situation as it is (baseline option) 38

4.3.2.       Policy option No 3/2: Facilitating GCP inspections by increasing transparency. 38

4.3.3.       Policy option No 3/3: Inspections of non-EU countries' regulatory systems for clinical trials  39

4.3.4.       Policy option No 3/4: GCP inspections of non-EU countries' clinical trial sites. 39

4.3.5.       Policy option No 3/5 — Combination of policy option No 3/2 and No 3/3. 39

5.     Impact of policy options. 39

5.1.      Objective No 1 — A modern regulatory framework for submission, assessment and regulatory follow-up of applications for clinical trials. 41

5.1.1.       Policy option No 1/1: No action at Union level and reliance on voluntary cooperation of Member States (baseline option) 41

5.1.2.       Policy option No 1/2 — Single submission with separate assessment 44

5.1.3.       Policy option No 1/3 — Single submission with joint assessment by Member States of issues not related to ethical aspects. 45

5.1.4.       Policy option No 1/4: Single submission with central assessment by the Agency of issues not related to ethics  48

5.1.5.       Policy option No 1/5 — Choice of legal form — Adopting the text of the Clinical Trials Directive in the form of a Regulation. 51

5.1.6.       Policy option No 1/6 — Combination of policy option No 1/3 and No 1/5. 52

5.1.7.       Comparison of policy options for objective No 1 and synergies. 52

5.2.      Objective No 2 — Regulatory requirements adapted to practical considerations and needs. 56

5.2.1.       Policy option No 2/1: No action at Union level (baseline option) 56

5.2.2.       Policy option No 2/2 — Enlarging the scope of non-interventional trials. 56

5.2.3.       Policy option No 2/3 — Excluding ‘non-commercial  sponsors’ 57

5.2.4.       Policy option No 2/4: Removing regulatory requirements on the basis of the knowledge of the IMP  58

5.2.5.       Policy option No 2/5 —Insurance/optional ‘national indemnification mechanism’ 60

5.2.6.       Policy option No 2/6 — Combination of policy options No 2/4 and No 2/5. 61

5.2.7.       Comparison of policy options for objective No 2 and synergies. 62

5.3.      Objective No 3: Addressing the global dimension of clinical trials when ensuring compliance with GCP  65

5.3.1.       Policy option No 3/1: Leaving the situation as it is (baseline option) 65

5.3.2.       Policy option No 3/2: Facilitating GCP inspections by increasing transparency. 65

5.3.3.       Policy option No 3/3: Inspections of non-EU countries' regulatory systems for clinical trial 66

5.3.4.       Policy option No 3/4: GCP inspections of non-EU countries' clinical trial sites. 66

5.3.5.       Policy option No 3/5: Combination of policy options No 3/2 and 3/3. 67

5.3.6.       Comparison of policy options for objective No 3 and synergies, subsidiarity. 67

6.     Conclusion — Final choices of policy options — Overview.. 69

6.1.      Final choices of policy options. 69

6.2.      Final overview.. 70

7.     Monitoring and evaluation. 71

Annex 1: Clinical Trials and the Clinical Trials Directive.

Annex 2: Key figures.

Annex 3: Objective No 1 — A modern regulatory framework for submission, assessment and regulatory follow-up of applications for clinical trials.

Annex 4: Objective No 2 — Regulatory requirements adapted to practical considerations and needs.

Annex 5: Objective No 3 — Addressing the global dimension of clinical trials when ensuring compliance with GCP 

Annex 6: Costs and financing strategies for the single submission point

Annex 7: Policy option 1/3: Support structure at EU level - implementation costs.

Annex 8: Involvement of Commission or Agency – Key points for consideration.

Acronyms

AESGP || Association of European Self-medication Industry

ASR || Annual Safety Report

CIOMS || Council for International Organisations of Medical Sciences

CoE || Council of Europe

CTFG || Clinical Trials Facilitation Group

EATG || European Aids Treatment Group

EC || Ethics Committee

ECPC || European Cancer Patient Coalition

EEA || European Economic Area

EFPIA || European Federation of pharmaceutical industries and associations

EPF || European Patient's Forum

EMA || European Medicines Agency (‘the Agency’)

EORTC || European Organisation for Research and Treatment of Cancer

EUCOPE || European Confederation of Pharmaceutical Entrepreneurs

EUCROF || European Contract Research Organisation Federation

EVCTM || Eudravigilance – Clinical Trials Module

FTE || Full-Time Equivalent

FVO || Food and Veterinary Office of the European Commission

GCP || Good Clinical Practice

IAB || Impact Assessment Board

ICH || International Conference on Harmonisation of Technical Requirements for Registration of Pharmaceuticals for Human Use

ICREL || (Study on) ‘Impact on Clinical Research of European Legislation’

IMP || Investigational Medicinal Product

MAA || Marketing Authorisation Application

MHRA || Medicines and Healthcare products Regulatory Agency of the UK

NCA || National Competent Authority

OECD || Organisation for Economic Cooperation and Development

PIP || Paediatrics Investigation Plan

PSUR || Periodic Safety Update Report

SA || Substantial Amendment

SME || Small and Medium-sized Enterprise

SUSAR || Suspected Unexpected Serious Adverse Reaction

TFEU || Treaty on the Functioning of the European Union

VHP || Voluntary Harmonised Procedure

1.           Procedural issues

1.           In its Communication of 10 December 2008 to the European Parliament, the Council, the European Economic and Social Committee and the Committee of the Regions on ‘Safe, Innovative and Accessible Medicines: a Renewed Vision for the Pharmaceutical Sector’,[1] ('the 2008 Pharmaceuticals Communication') the Commission announced that an assessment would be made of the working of Directive 2001/20/EC of the European Parliament and of the Council of 4 April 2001 on the approximation of the laws, regulations and administrative provisions of the Member States relating to the implementation of good clinical practice in the conduct of clinical trials on medicinal products for human use[2] (the ‘Clinical Trials Directive’).

2.           This assessment would consider, in particular, various options for improving the functioning of the Clinical Trials Directive with a view to making legislative proposals, if appropriate, while taking the global dimension of clinical trials into account.

3.           This impact assessment and adoption of the legislative proposal in 2012 are contained in the Commission work programme for 2011[3] and 2012[4] and are scheduled in the Commission’s ‘agenda planning’ under reference number 2011/SANCO/015.

1.1. 1.1.    Consultations of other Commission departments and agencies

4.           An Inter-Service Steering Group was set up which met on various occasions. The meetings were attended by representatives from Directorates-General ENTR, DEVCO, JUST, RTD, BUDG and the Secretariat-General. Close contacts were maintained with the European Medicines Agency (‘the Agency’) on this file.

1.2. 1.2.    Consultations of Member States

5.           The proposed revision of the Clinical Trials Directive was presented to and discussed by the Pharmaceutical Committee[5] at its 66th (14 February 2011) and 67th (5 October 2011) meeting.

6.           Various technical aspects of the impact assessment were discussed with Member States’ representatives at the meetings of the ‘Ad hoc group for the development of implementing guidelines for the ‘Clinical Trials Directive’ 2001/20/EC’[6].

1.3. 1.3.    Stakeholder consultations

7.           In October 2007, the Commission, jointly with the Agency, held a one-day conference on ‘Operation of the Clinical Trials Directive (Directive 2001/20/EC) and Perspectives for the Future’ (hereinafter referred to as the ‘Commission/Agency clinical trials conference’). The results of that conference were published in an extensive report.[7]

8.           The Commission held a stakeholder consultation from 9 October 2009 to 8 January 2010 on the basis of a public consultation document (hereinafter referred to as the ‘2009/10 public consultation’).

9.           This stakeholder consultation was followed up by a public consultation on a concept paper concerning revision of the Clinical Trials Directive (hereinafter referred to as the ‘2011 public consultation’).[8] This public consultation was open from 9 February to 13 May 2011.

10.         Topics which had been explored extensively during the first consultation were not put forward again for discussion. Instead, the purpose of the 2011 public consultation was:

· to seek more specific ideas on the issues that were presented in a rather general way during the 2009/10 public consultation. Consequently, some issues considered in the concept paper were of a more detailed and technical nature; and

· to verify with stakeholders the core data which form the basis of the impact assessment.

11.         Thus, the concept paper submitted in the 2011 public consultation was more detailed and specific. It presented:

· a ‘preliminary appraisal’ of the options which appear to be the most suitable to address some of the key concerns of the Clinical Trials Directive, on the basis of the current state of the impact assessment; and

· the main figures that are being used to evaluate the impact of the various policy options.

12.         In both public consultations, all the ‘General principles and minimum standards for consultation of interested parties by the Commission’[9] were met. The responses, and a summary of them, have been published by the Commission.[10] In addition, the main results of the public consultations are taken up, grouped according to stakeholder groups, throughout this report. A general appraisal of the various stakeholder groups is contained in Annex 1.

13.         In addition, the Commission held several meetings with stakeholders to hear their assessment of how the Clinical Trials Directive is working and to discuss the impact of potential policy options. A first round of meetings was held with stakeholder groups (patients, industry and academic researchers) in 2009. In the course of the 2011 public consultation a large stakeholder workshop was held on 31 March 2011 to clarify various points put forward in the concept paper. Moreover, this workshop gave stakeholders an opportunity to discuss their concerns together and to get to know each other’s views.

14.         Finally, both in the run-up to and throughout the impact assessment process, stakeholders launched several projects and published the results in several documents. They include:

· The recommendations of the High-Level Group of Independent Stakeholders on Administrative Burdens (‘Stoiber Group’) of 5 March 2009;[11]

· The ‘forward look’ by the European Science Foundation on ‘Investigator-driven clinical trials’,[12] published in March 2009;

· The ‘Road Map Initiative for Clinical Research in Europe’ of the multi-stakeholder 'European Forum for Good Clinical Practice'. In the context of this road map initiative a series of workshops were held which concluded with suggestions as to how to improve the legislation on clinical research;[13]

· The project ‘PatientPartner — Identifying the Needs of Patients Partnering in Clinical Research’.[14]

15.         Furthermore, the OECD has launched a working group in order to explore how to facilitate multinational cooperation in investigator-driven clinical trials.[15]

16.         The Commission participated actively either in the projects themselves or in follow-up conferences and workshops.

1.4. 1.4.    Contacts with non-EU authorities

17.         In the course of the impact assessment, the Commission has been in close contact with the US Institute of Health. There have also been contacts with the authorities of several other non-EU countries (including Japan, China and India).

1.5. 1.5.    Impact Assessment Board

18.         The impact assessment was submitted to the Impact Assessment Board (IAB) for scrutiny.[16] In its first opinion (which is publicly available on the EUROPA server[17]), the IAB requested the following improvements of the draft impact assessment report:

           Providing a clearer and more concise problem description, including a better presentation of stakeholder views, a presentation of the enterprises and research bodies primarily affected by the situation, concrete examples, and a discussion of the causality between the regulatory framework for clinical trials in the EU, and the decline of clinical trials conducted in Europe. Moreover, the IAB requested to explain the relationship between medicines legislation and clinical trial regulation. These aspects have been addressed in section 2 of the report.

           Strengthening the 'intervention logic', by introducing operational objectives and better linking the policy options to them. Moreover the different policy options should be explained in more detail: This has been taken up in section 3 and in the presentation of the policy options in section 4 of the report.

           Better presenting the impacts of the policy options, in particular by presenting stakeholder views on the policy options and by addressing combination of policy options separately and comparing them against the baseline. To address this, throughout section 5 of the report a short summary of stakeholder viewpoints, according to stakeholder groups, has been added.

           Better explanation of monitoring and evaluation arrangements: This is addressed in section 7 of the report.

19.         In its second opinion, the IAB requested the following improvements of the draft impact assessment report:

           A better link between the the problems experienced by sponsors and investigators, and the Clinical trials Directive. To address this, the report has been amended in sections 2 and 4.

           A better explanation of the policy option No 2/5 ('national indemnification mechanism'. This has been addressed in the respective description of the policy option in section 5 of the report.

           A clearer outline of the evaluation arrangements. This has been addressed in section 7 of the report.

20.         Moreover, in its second opinion the IAB requested to shorten the report, for example by moving parts of the problem description into the Annexes. To respond to this request, the problem description has been shortened, and Annex 1 of the report has been amended.

2.           Problem definition

1.6. 2.1.    Introduction — setting the scene

1.6.1. 2.1.1. What are clinical trials?

21.         A clinical trial as defined in the Clinical Trials Directive is ‘any investigation in human subjects intended to discover or verify the clinical, pharmacological and/or other pharmacodynamic effects of one or more investigational medicinal product(s), and/or to identify any adverse reactions to one or more investigational medicinal product(s) and/or to study absorption, distribution, metabolism and excretion of one or more investigational medicinal product(s) with the object of ascertaining its (their) safety and/or efficacy’.[18]

22.         Clinical trials are performed in many different contexts: applications for marketing authorisation for medicinal products contain large amounts of data generated in clinical trials. Publications in medical journals are also often based on data generated in clinical trials. Therefore, clinical trials are an indispensable part of clinical research which, in turn, is essential to develop medicinal products and improve medical treatment. Without clinical trials, there would be no new medicines, no further development of existing medicines, and no evidence-based improvement of treatments with medicines.

23.         In the EU/EEA,[19] approximately 4 400 clinical trials are applied for every year. This equals approximately 10 000 applications in the Member States (one clinical trial can mean up to 27 clinical trial applications, see Annex 2, tables 2 and 3). Approximately 60 % of clinical trials are sponsored[20] by the pharmaceutical industry and 40 % by other stakeholders, such as academics.[21] They aim to improve and optimise the use of authorised medicines, but could also well be done with the intention of developing a medicinal product. Detailed figures on clinical trials in the EU are given in Annex 2.

24.         Approximately 24% of all clinical trials applied for in the EU are multinational clinical trials, i.e. clinical trials intended to be performed in at least two Member State. While this seems a relatively small proportion, it has to be highlighted that these 24% clinical trials involve approximately 67% of all subjects enrolled in a clinical trial.

25.         This means that, in average, a clinical trial with more than 40 subjects is conducted in more than one Member State. Mono-national clinical trials are limited to small studies with low recruitment targets.

26.         Having said this, multi-national clinical trials do not necessarily involve all Member States. Rather, in practice, multi-national clinical trials are only rarely being performed in more than 6-8 Member States. For example, in 2010, of the 4 400 clinical trials applied for in the EU, only 268 (approximately 6 %) were to be rolled out in eight Member States or more (see Annex 2).

             

Chart: Share of multinational clinical trials applied for in the EU in 2010

             

Chart: Share of subjects in multinational clinical trials applied for in the EU in 2010

27.         The risk to safety of patients participating in a clinical trial depends on a variety of factors, in particular the extent of knowledge of the investigational medicinal product ('IMP') and the type of intervention in the trial. At one end of the spectrum are ‘first-in-man’ (Phase I) clinical trials[22] with compounds previously not administered to humans. At the other are clinical trials with well-known medicines which are used in the authorised indication, or one very similar, and where additional interventions do not go much beyond normal clinical practice (e.g. an additional blood sample or questionnaire). One example of these low-risk trials are large, randomised treatment optimisation studies, where authorised medicines are used in a clinical trial setting in order to improve standard therapies (see point 2.2.2).

1.1.1. The regulation of clinical trials in the EU

28.         Clinical trials are regulated by the Clinical Trials Directive. The key aim of the Directive is compliance with good clinical practice (GCP). According to the Clinical Trials Directive, ‘good clinical practice is a set of internationally recognised ethical and scientific quality requirements which must be observed for designing, conducting, recording and reporting clinical trials that involve the participation of human subjects. Compliance with this good practice provides assurance that the rights, safety and well-being of trial subjects are protected, and that the results of the clinical trials are credible.’[23] A description of the main contents of the Clinical Trials Directive is contained in Annex 1.

29.         The conditions for conducting clinical trials, and in particular their authorisation, are unrelated to the regulation and authorisation of medicinal products:

           The Union acquis on medicinal products addresses the question whether and under what conditions a medicinal product can be placed on the EU market.

           The aim of this legislation is to ensure that medicinal products placed on the market are of high quality as well as a favourable benefit-risk balance of the product. In this context, the Union acquis on medicinal products provides for various types of advise to (future) marketing authorisation holders as to what clinical data is desirable from the point of view of a marketing authorisation.

           The Union acquis on clinical trials addresses the question whether and under what conditions a clinical trial can be performed in the EU (a clinical trial may be conducted with medicines already authorised (see point 2.1.1), or with medicines not yet authorised).

           The aim of this legislation is to ensure the rights and safety of the subject, and to ensure that the data generated in a clinical trial is reliable and robust (for example, in view of statistical methods used, and in view of the endpoints measured). Thus, this legislation sets out what clinical trial is acceptable in view of the risk to subject rights and safety, and in view of the reliability and robustness of the data generated.

30.         Thus, the authorisation of a medicinal product and the authorisation of a clinical trial follow different aims. These aims may even be in conflict: The conduct of a clinical trial may be desirable from the point of view of authorisation of a medicinal product, while from the point of view of the conduct of clinical trials, it cannot be authorised in view of subject rights and safety.

31.         The interface between the EU legislation on medicinal products and the EU legislation on clinical trials is limited to one aspect: the acceptability of clinical data in the marketing authorisation process. The clinical data submitted with a request for a marketing authorisation for a medicinal product in the EU has to stem from clinical trials conducted in accordance with the Clinical Trials Directive. If the clinical data stems from clinical trials conducted outside the EU, the clinical trial has to be conducted on the basis of principles which are equivalent to those applied in the Clinical Trials Directive.[24]

1.1.2. Affected bodies and enterprises – the "sponsor"

32.         The Clinical Trials Directive establishes the notion of 'sponsor', which is the person responsible for the clinical trial. Broadly speaking, two types of sponsors can be identified:

           Pharmaceutical companies ('industry sponsors'): These range from large multinational research-based pharmaceutical companies to small, research-based pharmaceutical companies (on the share of small and medium enterprises, see point 5).

           'Non-commercial sponsors':[25] Although the term 'non-commercial sponsor' is not defined anywhere, it is generally understood to mean sponsors of clinical trials whose results are not intended to be used, prima facie, for authorisation or development of a medicinal product or for further extension of a medicinal product to other therapeutic areas. ‘Non-commercial sponsors’ are usually universities or academic institutes, foundations or charities. 'Non-commercial sponsors' range from large research organisations with well-organised structures to small, fragmented cooperative structures with a far lower level of dedicated resources.

33.         These two types of sponsors are often interlinked: for example, research organisations may carry out clinical trials for pharmaceutical companies and academic research may, through publications, influence the development of medicinal products.

1.6.2. 2.1.4. The benefits of clinical trials – Public and patient health

34.         The conduct of clinical trials is beneficial for innovation in public health and patient health, and brings about important investments into the healthcare sector.

Public health:

35.         Clinical trials allow for improving public health in the EU and worldwide. Both the big advances and small, incremental improvements in public health in the last decades were possible largely thanks to clinical trials.

Example: ISIS-2 (the International Study of Infarct Survival) tested a new approach using aspirin and streptokinase in combination immediately after a heart attack. Use of either streptokinase or aspirin alone reduced the risk of vascular events by 25 per cent, but the two together decreased the risk by 50 per cent.[26]

36.         Apart from breakthrough studies like this, continuous advances in treatment have contributed to significant improvements in public health, not only in the EU but also worldwide. This progress was possible largely thanks to clinical trials. An example of this progress is the improvement in cancer survival rates in the EU over the last few decades.[27]

Chart - Example: Survival rates of children and young adults suffering from cancer

Patient health:

37.         Apart from these societal benefits at large, clinical trials can be beneficial for patient health in the concrete setting of a clinical trial. Indeed, patients, in particular patients who suffer from serious or life-threatening diseases, are often keen to participate in clinical trials, as this may be the only way to access a treatment. The reasons why patient access to a medicine or to a medical treatment may only be possible through participation in a clinical trial are manifold: the medicinal product may not be available in a given Member State outside a clinical trial, or not be re-imbursed. Moreover, new medical treatments may not yet be widely spread and only be applied in the context of a clinical trial.

38.         Therefore, patient organisations privately run 'clinical trial registries' in order to allow citizens to take part in research which may improve their conditions. Partly in order to respond to this urgent demand, the legislator has, in 2004, provided that information on clinical trials is to be made publicly available in the EudraPharm database.[28]

39.         Finally, while still much under debate, there are also studies suggesting that the participation in clinical trials is beneficial for the patient itself independently from whether the patient takes part in the 'experimental arm' or in the 'control arm' of a clinical trial.[29]

Investments:

40.         Conducting clinical trials entails considerable investment and growth in the EU, including inward investment by sponsors from non-EU countries (see Annex 2 for details). In recent years a range of publications have highlighted these tangible benefits of clinical trials.[30]

1.7. 2.2.    Problem identification

Introduction

41.         The Clinical Trials Directive has brought about important improvements in the safety and ethical soundness of clinical trials in the EU and in the reliability of clinical trials data. This has been confirmed in numerous fora. For example, at the Commission/Agency clinical trials conference a large majority of the delegates acknowledged that, overall, the Directive had resulted in better protection of participants in clinical trials.[31]

42.         Moreover, the Directive has led to cooperation in this area between Member States, who now meet regularly in three settings: the ‘Ad hoc group on implementation of the ‘Clinical Trials Directive’ 2001/20/EC’ (organised and chaired by the Commission), the inter-governmental ‘Clinical Trials Facilitation Group’ ('CTFG' - organised and chaired by Member States) and the ‘GCP Inspectors Working Group’ (organised and chaired by the Agency).

43.         Nevertheless, the Clinical Trials Directive is the most heavily criticised piece of legislation of the entire EU acquis for pharmaceuticals. The criticism focuses on a too cumbersome and bureaucratic regulatory framework in the EU, which did not come along with a genuine harmonisation of administrative requirements. The severe criticism is voiced by all stakeholders and political actors ‑ patients, industry, and academic research, Member States, Union institutions ‑ and has been re-iterated and stressed during the various consultations referred to in point 1.3. Annex 1 sets out the broad criticism voiced by each of these actors and stakeholders:

44.         The negative consequences of the fragmentation of the authorisation procedure were also highlighted by the High-Level Group of Independent Stakeholders on Administrative Burdens (‘Stoiber Group’) in its recommendations of 5 March 2009.[32]

45.         These criticisms are supported by the data available:

46.         Decrease in the number of applications for clinical trials: According to the official EU-database for clinical trials (EudraCT), since 2007 the number of clinical trials applied for in the EU has fallen by 12% to 4 400 clinical trials applied for in 2010.

Chart: Clinical trials applied for in the EU[33]

             

47.         Increased costs for conducting clinical trials: In order to do a retrospective assessment of the effects of the Clinical Trials Directive, the Commission launched in 2008 a comprehensive study on the ‘Impact on Clinical Research of European Legislation’ (ICREL)[34] as part of the 7th Framework Programme. ICREL showed the following:

           Compared to the situation prior to the application of the Clinical Trials Directive, the staff needs for industry sponsors to handle the clinical trial authorisation process have doubled (107%); Small companies faced and even stronger increase. In some areas of clinical trial regulation, such as safety reporting, the number of FTEs in pharmaceutical companies increased by 85%;[35]

           Regarding non-commercial sponsors the increase of administrative requirements due to the Clinical Trials Directive has lead to an increase of costs in administration of 98%;[36]

           Since implementation of the Clinical Trials Directive the insurance fees have dramatically increased for industry sponsors by 800%.[37]

48.         Delays for launching a clinical trial: the average delay between finalisation of the protocol and the ‘first patient in’ has increased by 90 % to 152 days[38].

49.         The ICREL findings are similar to data published in numerous articles: In Hearn and colleagues[39] the authors investigated the impact of the Clinical Trials Directive on eight clinical trials units in UK. Results show that costs have doubled, the start of the trials was delayed and starting and conducting trials was much more difficult than before. As for the clinical research activity, Moulton[40] reported a decrease of 25% in submissions in Sweden; 40% in Ireland with a drop of 60% from non-commercial sponsors. The same was found by the Cancer Research UK where the number of clinical trial applications was down by approximately 50%. The European Organisation for Research and Treatment of Cancer (EORTC)[41] faced the same situation: from 23 new studies in 2003 to 10 in 2007.

50.         Despite this data one has to assess whether the decline is really caused by the Clinical Trials Directive, or by other causes than regulation. An assessment of possible other causes reveals, however, that the Clinical Trials Directive is an important direct or indirect driver of the numbers of clinical trials in the EU.

51.         The following factors may contribute to the decline clinical trials in the EU:

· Industry conducts, generally, less research: The economic slowdown in Europe since 2009/10 might be considered as a cause for reduced clinical trials activity in the EU. However, nothing indicates that the pharmaceutical industry, overall, conducts less research than before: On the contrary, the constant increase of the amount of data requested in the context of a marketing authorisation (see point 2.2.1.4) leads, overall, to more research conducted by pharmaceutical companies themselves, or on their behalf. However, there is a general assumption that clinical research is increasingly conducted outside the EU (see below under this point).

· Less public funding available: Regarding non-industry research, one may argue that, in view of the macroeconomic climate in Europe since 2010, there is less public funding available for conducting academic research. Even if this was the case, however, the effect of these cuts would not yet be reflected in the figures above.

· More difficulties to recruit patients: Increasingly narrowly-defined disease profiles (see point 2.2.1.3) have made it more challenging to recruit patients. In practice, therefore, today, every larger clinical trial takes place in more than one Member State (see point 2.1.1). However, it is precisely these clinical trials that are particularly challenging in terms of clinical trials regulation in the EU (see point 2.2.1).

· Increasing costs in terms of salaries, hospital service fees, etc.: One may argue that the increased costs for staff in the health sector makes it more difficult to conduct clinical trials in the EU. However, as far as clinical research is concerned, these costs are to a considerable extent influenced by regulatory requirements, and in particular the Clinical Trials Directive (see baseline option).

52.         In conclusion, it would be wrong to attribute the decline of clinical trial activity solely and exclusively to the Clinical Trials Directive. However, the Clinical Trials Directive has had many direct effects on costs and feasibility of conducting clinical trials which, in turn, lead to a decline of clinical trial activity in the EU. Moreover, other causes (such as salary costs and the need to conduct multinational studies in order to reach recruitment targets) are aggravated through regulatory requirements and consequential costs of the Clinical Trials Directive.

53.         This raises the question whether the Clinical Trials Directive has simply "stopped" clinical trials, or whether such research is taken to non-EU countries. In this respect, a distinction has to be drawn between industry sponsors and non-commercial sponsors:

· Regarding industry sponsors, there is a trend towards globalisation of the conduct of clinical trials which are increasingly conducted in emerging economies such as India, China and various South American states, as well as Russia.[42] Various studies, as well as the media,[43] have highlighted a "dramatic shift"[44] in the location of trials from the traditional trial regions (North America and Europe) to new, emerging economies in the last years. While there is no ultimate proof available for the causality, all available sources suggest that, indeed, the reduced attractiveness of the 'traditional' trial countries in terms of costs contribute to the globalisation of the conduct of clinical trials. This effect has been highlighted in a number of scientific publications, and in both public consultations conducted by the Commission. For example, the Belgian pharmaceutical industry association 'Pharma.be' stressed that "The emerging countries are attracting a growing number of large-scale clinical trials as they have access to large patient populations required to run these trials. […] This shift to the 'rest of the world' has increased markedly in recent years and the trend looks set to continue, ultimately leading to a drop in clinical research activities in EU and US. […] Creating a regulatory framework that favours the conduct of clinical trials at EU level should by all means be reinforced to keep clinical research in Europe."[45]

· Regarding non-commercial sponsors, the situation is different as these actors usually do not have access to the globalised clinical trial market. In the case of non-commercial sponsors, if the regulatory and other impediments are too high, these clinical trials are simply not being performed.

54.         The concrete problems which are thus to be addressed are the following:

1.1.3. Separate submission, diverging assessments and regulatory follow-up of applications for clinical trials

1.7.1.1. 2.2.1.1.          The issue

55.         The Clinical Trials Directive provides that a clinical trial, prior to it being conducted has to be authorised by two distinct bodies: the national competent authority (NCA) and one or more Ethics Committee(s) (EC).[46] The "authorisation"[47] consists of two steps, the "submission" and the "assessment" (with a subsequent decision). In the assessment stage, the documentation in relation to the clinical trial is assessed in order to check compliance with the regulatory requirements of the Clinical Trials Directive (for more details on these requirements, see Annex 1). The various aspects are being checked either by the NCA, or by the EC, or by both, depending on national practices and traditions.

56.         The delay for authorising a clinical trial is, according to the Clinical Trials Directive, 60 days, subject to some exceptions.[48]

57.         In addition, clinical trials are subject to regulatory follow-up/supervision. This includes any subsequent changes to the clinical trial (‘substantial amendments’ — SA), safety information, end-of-trial declarations, etc.

58.         The submission, assessment and regulatory follow-up for the same clinical trial are conducted in the different Member States completely separately from one another. Thus, while the Clinical Trials Directive introduces a submission/assessment/authorisation process, it does not provide for any kind of cooperation or exchange of information. Neither does the Clinical Trials Directive give, in this process, any role to the Commission or to the Agency,[49] i.e. the entire process of submission, assessment and follow-up is conducted without any involvement of Union institutions or bodies.

59.         The authorisation process for clinical trial, introduced in the EU with the Clinical Trials Directive, has been a key criticism of all stakeholders in the last years. This criticism has been voiced since the adoption of the Clinical Trials Directive. It has been highlighted in particular in the Commission/Agency clinical trials conference,[50] in the 2009/10 public consultation, where in particular the negative effect on SMEs was stressed.[51] The delays of the actual full launch of the trial and increase in costs as established with ICREL (see points 2.2 and 2.2.1.2) are to a large extent attributed to the current system of submission, assessment and regulatory follow-up of clinical trials.

Chart: Submission/assessment procedure today (example: four Member States)

Separate submissions

60.         For any clinical trial, the information required for the authorisation is submitted to each Member State separately. Moreover, usually, within each Member State the information is submitted to the two assessment bodies, the NCA and the EC, separately.

Separate assessments

61.         As mentioned earlier, each clinical trial is subject to an assessment by two distinct bodies, the NCA and the EC of each Member State concerned. The scope of their assessments differs in each Member State, depending on national traditions and expertise. This renders even voluntary cooperation between Member States more difficult and further complicates authorisation of clinical trials in the Union. Furthermore, the requirements set out in the Directive for the assessment are applied very differently in the individual Member States concerned

62.         In this context, it must be stressed that while the outcome of the assessment (i.e. clinical trial is approved or not) usually does not differ[52], there are many differing, and sometimes conflicting, requests for additional information or national changes to a protocol.[53]

63.         In the 2011 public consultation it was confirmed that in approximately 80 % of all multinational clinical trials, the feedback from the NCAs diverges as regards:

- requests for additional information; or

- Grounds for non-acceptance.[54]

Separate regulatory follow-up/supervision

64.         The difficulties are also echoed throughout the entire regulatory follow-up of a clinical trial. This includes any subsequent important changes to the clinical trial (‘substantial amendments’ — SA), safety information, end-of-trial declarations, etc.

65.         In particular, concerning SAs, any change to the documentation submitted which is ‘substantial’ is subject to approval by each Member State individually. Within each Member State this is given either by the NCA or by the EC or by both. Again, this creates divergences in the regulatory assessment of a clinical trial. The difficulties described above with ‘separate assessment’ are repeated.

1.7.1.2. 2.2.1.2.          Consequences

· Costs: The administrative burden and administrative costs for sponsors increase, without any added value: in both public consultations stakeholders submitted concrete evidence through examples that resources are moved from research to bureaucracy. For example, it was reported that, for a single clinical trial with 280 participating clinical trial sites, 100.000 copied pages of documents had to be submitted to various authorisation bodies. In another study, 12.000 pages of documents had to be provided for a study conducted at 13 trial sites.[55] This is a direct result of the concept of 'separate submissions' referred to under point 2.2.1.1.

· Delays: The actual full launch of a clinical trial is delayed. As set out above (point 2.2.1.1), assessments of a clinical trial application very rarely conclude with simple "go"/"no-go": rather, the assessment concludes with additional questions, comments or conditions. These come in, from the various Member States concerned, at different time points and require adjustments to the planned clinical trial. In order to ensure data reliability and robustness, however, the sponsor is forced to maintain, as far as possible, one protocol. This leads to constant updates, amendments and changes of the documentation in different Member States during the assessment phase or right thereafter before a clinical trial can be rolled out. This "chain reaction"[56] was frequently highlighted in the 2009/10 public consultation[57] as a key cause for the long delays of the actual start of the trial as evidenced in the ICREL study (see point 2.2).

66.         Exclusion of Member States: Divergencies in the assessments can lead to contradictory decisions on the requirements for the same clinical trial. As a result, sponsors decide not to perform the clinical trial in a given Member State. This point was highlighted in the 2009/10 public consultation.[58] This means that access to innovative, potentially life-saving, treatment is denied to patients in one Member State to the advantage of those in another.

67.         The detailed assessment of the consequences is set out in detail in the description of the ‘baseline’ below (policy option No 1/1 — no action at EU level).

1.7.1.3. 2.2.1.3.          Link with the Clinical Trials Directive

68.         The consequences set out under point 2.2.1.2 are a direct consequence of the Clinical Trials Directive: This Directive introduced a multiple submission/assessment process for mono-national as well as multi-national clinical trials. At the same time, the Clinical Trials Directive did not introduce any form of cooperation or coordination in these multiple assessments.

1.7.1.4. 2.2.1.4.          Outlook

69.         The responses in both public consultations highlighted that the problems described, and the consequences, are going to worsen in the future. While, today, approximately 25% of all clinical trials are being performed in more than one Member State, this share is going to increase further in the future due to the following developments:

· Diseases are increasingly narrowly defined and often linked to genetic characteristics of the subject (often described as the trend towards ‘personalised medicines’). In order to recruit sufficient subjects with these specific characteristics, sponsors have to roll out the clinical trial in several Member States. This holds particularly true for research into oncology;

· Increasingly, there is a need for research on specific patient populations, such as children, adolescents or the elderly. In practice it is sometimes difficult to recruit enough subjects from such specific patient populations. In order to meet recruitment targets, it is necessary to run multinational trials;

· The requirements and expectations of regulators and the research community for well-powered trials are constantly increasing. In order to power a clinical trial sufficiently, it has to be rolled out in several Member States.

1.1.4. Greater difficulties with conducting clinical trials due to regulatory requirements not adapted to practical considerations and needs

70.         Regulation of clinical trials addresses two distinct risks: the risk to patient safety and the risk to data reliability. The former can vary widely, depending on a range of factors, in particular:

· The extent of knowledge and prior experience with the IMP (in particular, whether or not the IMP is already authorised in the EU or elsewhere); and

· The type of intervention (which can range from a simple blood sample to a sophisticated biopsy).

71.         However, the Clinical Trials Directive does not sufficiently address these differences in risk and take them into account. Instead, the obligations and restrictions laid down in the Directive apply largely irrespectively of the risk to subject safety and without matching practical considerations and requirements.

Example: A clinical trial comparing the efficacy of two authorised medicines (A) and (B), which are both used in their authorised indication. There is no certainty about the best treatment choice. The additional intervention is limited to randomisation of the patients (some receive medicine A; others receive medicine B), and to an additional standard intervention (e.g. additional measurements of blood pressure). Considering that the patient would have received medicine A or medicine B anyway, this clinical trial poses no additional risk compared to normal clinical practice.

72.         The disproportionate burden imposed by the Clinical Trials Directive is most obvious in the case of two key regulatory requirements in the Directive:

· Obligatory insurance/indemnity: Under the Clinical Trials Directive, the liability of the investigator or sponsor for possible injury or death of a participant in the clinical trial has to be covered by insurance or indemnity.[59] Thus, the Clinical Trials Directive has provoked the following situation:

· Insurance/indemnity is obligatory, i.e. sponsors/investigators are forced to obtain insurance coverage on the insurance market;

· This obligation applies to a small market: There are, at any given moment, approximately 12000 clinical trials ongoing in the EU. This is a very small segment in the insurance market for liabilities, in particular when comparing with other segments such as general liability insurance or automobile insurance.

The combination of a small market and an obligatory insurance (introduced with the Clinical Trials Directive) is the cause for a strong increase of the costs for premiums. ICREL has shown that, since implementation of the Clinical Trials Directive, the insurance premiums have increased for industry sponsors by 800%[60] even though justified claims continued to be are very rare and the compensation payments low. This 'aberrantly expensive'[61] insurance/indemnification coverage creates a disincentive to conduct clinical trials in the EU. This is discussed in more detail in the baseline option below (point 5.2.1.2).

· Obligatory annual safety report in the context of pharmacovigilance: Under the Clinical Trials Directive, every year the sponsor has to draft an ‘annual safety report’ (ASR) for every clinical trial.[62] The annual safety report has an equivalent for medicines which are authorised and used outside the context of a clinical trial: the ‘periodic safety update report’ (PSUR). The ASR creates considerable costs for sponsors. The actual costs are presented in the baseline option below (point 5.2.1).

73.         Both obligations apply independently of the actual risk which a clinical trial poses to the subjects. However, as mentioned above, these risks differ widely.

74.         This undifferentiated approach to regulation in the Clinical Trials Directive has been a key criticism of all stakeholders since the adoption of the Clinical Trials Directive in 2001. It has been a particular criticism of 'non-commercial sponsors' (see below). This criticism has been highlighted in the Commission/Agency clinical trials conference.[63] It was also stressed in both the 2009/10 and the 2011 public consultations, as well as in academic publications,[64] where stakeholders highlighted these as a heavy, and in many cases disproportionate, burden with associated increase in costs (including administrative burdens). Indeed, the increase in costs for administrative requirements as established with ICREL (see point 2.2) is – apart from the authorisation process ‑ to a large extent attributed to the administrative requirements in the Clinical Trials Directive which are not proportionate to the additional risk to a patient posed by a clinical trial.

75.         The issue set out in thie problem description is critical in particular for ‘non-commercial sponsors’ (see point 2.1.2) which have greater difficulties than industry sponsors to comply with the obligations set out in the Clinical Trials Directive in terms of budgetary and human resources. Indeed, practically all non-commercial sponsors have, since the adoption of the Clinical Trials Directive, heavily criticised this new regulatory framework for having hampered the conduct of clinical. In particular, the new regulatory requirements were heavily criticised. It was stressed that non-commercial sponsors could not comply with these requirements in view of the limited financial and human resources for compliance with regulatory requirements.

76.         In addition, it is noteworthy that this problem is also voiced by patients and patient groups, i.e. the stakeholders in whose interest clinical trials are actually regulated. This can be explained with patients' awareness of the benefits of clinical trials for evidence-based improvements of treatments, and possibly the benefits for individual patients participating in the trial (see point 2.1.4).

1.1.5. Reliability of clinical trial data in a globalised research environment

77.         Clinical trials are performed in the EU and in non-EU countries. About 25 % of all clinical trials performed in the EU also involve at least one non-EU country.

78.         As regards clinical trial data submitted in EU-wide marketing authorisation, 65 % of all data on patients submitted in pivotal clinical studies are generated in non-EU countries (see Annex 2).

79.         This trend towards globalisation of clinical research is expected to increase further in the next years.[65]

80.         Globalisation of clinical research, including in low-income non-EU countries, is by no means negative. Clinical research on a global scale is of benefit to the countries participating, to their populations and to global public health.

81.         However, despite universal agreement on the applicable principles of GCP (see Annex 1), the globalisation of clinical research poses a challenge when it comes to supervision of compliance with GCP. Any clinical trial which is referred to in the EU in the context of another clinical trial or of an application for marketing authorisation has to comply with GCP if it is to be considered reliable. In the case of data submitted in an application for marketing authorisation, Union legislation requires that clinical trials performed in non-EU countries have to be conducted on the basis of principles which are equivalent to those applied in the Clinical Trials Directive.[66] This may be checked, in the framework of a marketing authorisation procedure, by the Agency (or national competent authorities) through inspections. These inspections may be conducted at any relevant site, such as the clinical trial site, or the sponsor site. In the case of a marketing authorisation procedure at EU-level, the Agency does not dispose over inspection capacities itself. Rather, these inspection capacities are provided voluntarily by NCAs.

82.         In this context, it has to be stressed that there are, to date, no reliable, quantifiable data on whether the degree of non-compliance with GCP is higher in non-EU countries than in the EU. In particular, while the number of GCP inspections by the EU in non-EU countries is low and the sample size very limited (see Annex 2), the findings in these EU inspections in sites in non-EU countries do not differ significantly from GCP inspections conducted within the EU.

83.         However, this matter is widely discussed in regulatory and political settings. For example, since 2008, more than one third of all Parliamentary questions to the Commission in relation to clinical trials addressed this issue[67] and civil society groups are heavily engaged in this subject.[68],[69],[70] This is not surprising, since it is evident that clinical trials performed in non-EU countries, whose results are used in the EU, are more difficult to supervise and control.

84.         Thus, due to the difficulties to supervise and control clinical studies performed in non-EU countries, there is a continuing risk that advances in health in the EU will be based on clinical research not complying with the international standards adopted to guarantee the reliability of the results and protection of the subjects.

1.8. 2.3.    Union powers and subsidiarity

85.         Union legislation on clinical trials is based on Article 114 of the Treaty on the Functioning of the European Union (TFEU). It aims at harmonising the regulatory framework for pharmaceutical products, including the authorisation of their placing on the market. Harmonised rules open up the possibility of referring to the results and findings of clinical trials in applications for an authorisation for placing a medicinal product on the Union market, including subsequent variations and extensions of the marketing authorisation. In regulating clinical trials, the Union exercises its shared competence in accordance with Article 4(2) of the TFEU.

86.         This is critically important in the case of clinical trials because practically every larger clinical trial is performed in more than one Member State (see point 2.1.1).

87.         An additional factor is that medicinal products intended for research and development trials are excluded from the Community Code for medicinal products for human use.[71] IMPs may have been produced in a different Member State from that where the clinical trial is conducted. Thus, these products do not benefit from the secondary Union law ensuring their free movement while maintaining a high level of protection of human health.

88.         Situations like this were dealt with unsatisfactorily until the Clinical Trials Directive came into force. The laws, regulations or administrative acts differed from one Member State to another. These differences between national laws forced marketing authorisation holders to adapt their applications for authorisation to place their medicinal product on the market. They also hindered distribution of these products. This had a direct effect on the completion and operation of the internal market.

89.         To address this issue, it was necessary to harmonise the rules in place on the internal market. It would not have been possible for each Member State individually to establish identical rules. The EU legislation on clinical trials attempts to meet this need. It lays down, at Union level, the rules of procedure to be complied with on, inter alia, authorisation and performance of clinical trials, including safety reporting and manufacturing and labelling of medicinal products used in a clinical trial. These rules are exhaustive, i.e. they are not ‘minimum standards’. Member States are not allowed to ‘add to’ these rules.

90.         Any changes made to these rules by Member States would conflict with the requirements of the Treaty, as only the Union can amend them.

91.         This assessment also applies to legal acts adopted on the double legal basis of Article 114 and Article 168(4)(c) TFEU. Article 168(4)(c) TFEU provides an additional legal basis which was introduced into primary EU law by the Lisbon Treaty. Article 168(4)(c) TFEU confirms that the Union legislator, in order to meet common safety concerns, can set high standards of quality and safety for medicinal products. Since the entry into force of the Lisbon Treaty, all secondary legislation in the area of pharmaceuticals was based on this 'double legal basis'.[72]

92.         Having said this, in the case of regulation of clinical trials, while Union law on clinical trials has to comply with the rights, freedoms and principles set out in the Charter of Fundamental Rights of the European Union[73], the Treaty sets limits as regards harmonisation of ethical aspects of authorisation and regulation of clinical trials. Ethical aspects relate, in particular, to the need to obtain ‘informed consent’ from the subject or the legal representative. Any medical intervention requires consent from the patient but this is particularly critical for a clinical trial. Indeed, irrespective of the risk which a clinical trial may pose to a patient, the mere fact that the treatment is part of an experiment renders it necessary — from an ethical viewpoint — to obtain the informed consent of the subject. Hence, apart from some general principles set out in the Clinical Trials Directive, the detailed aspects of informed consent are of an ethical nature and intrinsically of national competence. Therefore they are not included in the scope of this harmonisation of regulation of clinical trials.

93.         There are also several aspects which are of an intrinsically national nature, in particular:

· Rules for establishing who is a ‘legal representative’ of a subject who cannot give informed consent (for example, because the subject is a child): these rules differ widely across the EU, depending on national tradition and practices;

· Rules on the extent of and prerequisites for liability for damages suffered by a subject: these rules are deeply rooted in national civil law on medical liability. This applies not only to the degree of negligence (e.g. no-fault or objective liability) but also to the rules on the burden of proof and for calculating the extent of damage.

94.         Consequently, while regulation of clinical trials and, in particular, revision of the existing Clinical Trials Directive, is compatible with the principle of subsidiarity, there are limits set by the Treaties which have to be considered when formulating the policy options.

3.           Objectives

95.         In accordance with the 2008 Pharmaceuticals Communication of the European Commission the general policy objective is to make the EU a more attractive place for conducting clinical trials by improving the regulatory framework for clinical trials in the EU, while taking into account the global dimension of clinical trials.

96.         More specifically, the following policy objectives shall be defined:

1.9. 3.1.    Objective No 1: A modern regulatory framework for submission, assessment and regulatory follow-up

97.         Objective No 1 shall be defined as 'a modern regulatory framework for submission, assessment and regulatory follow-up of applications for clinical trials, taking into account the multinational research environment'.

98.         In terms of operational objectives, this means the following:

· Reducing those administrative costs[74] which are 'administrative burdens',[75] and reducing other compliance costs;[76]

· Reducing the delay between finalisation of the protocol and the start of the trial, as far as this delay is caused by regulatory impediments.

1.10. 3.2.    Objective No 2: Regulatory requirements which are adapted to practical considerations, constraints and needs, without compromising the safety, well-being and rights of participants in clinical trials and without compromising data robustness

99.         Objective No 2 shall be defined as 'regulatory requirements which are adapted to practical considerations, constraints and needs, without compromising the safety, well-being and rights of participants in clinical trials and without compromising data robustness'.

100.       In terms of operational objective, this means reduction of the administrative burden and other compliance costs created by two key regulatory requirements in the Clinical Trials Directive: The annual safety report and the obligatory insurance/indemnity (see point 2.2.2). This operational objective targets in particular non-commercial sponsors who do not have access to the same (human and financial) resources as industry sponsors (see point 2.2.2).

1.11. 3.3.    Objective No 3: Addressing the global dimension of clinical trials when ensuring compliance with GCP

101.       In terms of operational objective, this means ensuring compliance with GCP of clinical trials conducted in non-EU countries but referred to in the EU in the context of another clinical trial or of an application for a marketing authorisation.

1.12. 3.4.    Coherence with strategic policy objectives of the EU and the Commission

102.       The general, specific and operational objectives can be seen as part of larger, strategic, policy objectives of the EU and the Commission. These include in particular:

· The objective of 'Smart growth – an economy based on knowledge and innovation': In its Communication 'Europe 2020 – a strategy for smart, sustainable and inclusive growth'[77] the Commission has called for strengthened research performance, stressing that 'Europe needs to focus on the impact and composition of research spending and to improve the conditions for private sector R&D in the EU';[78]

· The objective to 'Reducing inequalities in health' and to 'base health policy on the best scientific evidence derived from sound data and information, and relevant research': In its White paper 'Together for health: A strategic approach for the EU 2008-2013'[79] the Commission committed to these objectives as part of a 'strategy based on shared health values' which was defined as one of the four fundamental EU actions on health;

· The objective of a 'Simplification of the regulatory environment' in the EU: In its Communication 'Implementing the Community Lisbon programme: A strategy for the simplification of the regulatory environment'[80] the Commission committed to a simplification strategy at EU level. The annual work programmes of the Commission contain a Simplification Rolling Programme. The 2011 work program includes the revision of the Clinical Trials Directive;[81]

· The objective of 'Reducing administrative burdens in the European Union': In its Communication 'Action Programme for Reducing Administrative Burdens in the European Union'[82] the Commission has called for a joint reduction target of administrative burdens, caused by EU and national legislation of 25%, stressing that pharmaceutical legislation should be a priority area of action.

Regarding the strategic policy objectives to simplify the regulatory environment and reduce administrative burdens the Commission has re-confirmed its ambition in its Communication of 2010 'Smart regulation in the European Union'.[83]

4.           Policy options

1.13. 4.1.    Objective No 1 — A modern regulatory framework for submission, assessment and regulatory follow-up of applications for clinical trials

1.1.6. Policy option No 1/1 — No action at Union level and reliance on voluntary cooperation of Member States (baseline option)

103.       In this policy option (see chart under point 2.2.1.1) no action would be taken at EU level.

104.       With regard to this option, it has to be highlighted that Member States have started, on a voluntary basis, to cooperate and jointly assess applications for authorisation of clinical trials under the ‘voluntary harmonised procedure’ (VHP). This procedure was set up by Member States without the involvement of the Commission or the Union co-legislators. It is based on voluntary parallel submission to all participating Member States of a dossier requesting authorisation of a clinical trial.[84] Once Member States have informally agreed on authorisation of the clinical trial, the dossier is re-submitted formally to each Member State. The impact of the VHP in view of the specific and operational objectives is discussed in point 5.1.1.

1.1.7. Policy option No 1/2 — Single submission with separate assessment

105.       This policy option would consist of central submission, via an IT gateway located at EU-level, and subsequent separate assessment by the EC and the NCA of each Member State concerned.

106.       Thus, this policy option would be limited to an IT-functionality. Nevertheless, a single submission, instead of the multiple submission process (see problem identification under point 2.2.1 and the presentation of the baseline under point 5.1.1) would reduce administrative burden and thus contribute to the operational objectives in specific objective No 1 (see point 3.1). The impact of this policy option is discussed below (point 5.1.2).

1.1.8. Policy option No 1/3 — Single submission with joint assessment by Member States of issues not related to ethical aspects

107.       This policy option would consist of central submission and subsequent joint assessment by the Member States where the clinical trial takes place. Apart from the reduction in administrative costs and burdens created by multiple submissions, a joint assessment of the clinical trials application would help to avoid diverging assessments and thus reduce further administrative burdens, other compliance costs, and delays of the launch of the trial (see point 3.1 and the discussion on the impact of this policy option in point 5.1.3).

108.       Under this policy option the dimensions of a clinical trial touching on ethical aspects, however, would remain within the ambit of each individual Member State (see point 2.3).

109.       Concerning the aspects assessed jointly, one Member State would take up the task of coordinating the input from all Member States concerned, and draft the assessment report. This 'reporting Member State' would be determined by the sponsor. It can be expected that, in practice, the sponsor is most likely to choose the Member State where the sponsor is established.

110.       Regarding the aspects assessed jointly, the conclusion would be binding for all Member States concerned, unless a Member States 'opts out' from these conclusions, and consequently from this clinical trial. This 'opt out' could only be invoked in the specifically qualified circumstance of a serious risk to subject safety based on significant differences in normal clinical practice in that Member State. As set out above (point 2.2.1.1), the cases where there would be a genuine disagreement about the acceptability of a clinical trial as such are rare. Therefore, 'opt-outs' are going to be the exception.

111.       At the end of the assessment phase, each Member State would issue a single authorisation decision covering both the non-ethical and ethical aspects. Thus, this policy option would consist of one 'integrated decision' of both, non-ethical and ethical aspects.

112.       In terms of timelines, sponsors could realistically expect to receive approval or non-approval of a clinical trial within 60 days of submission of the dossier.

113.       The assessment would be made only by the Member States concerned and the conclusions would be valid for them only. If, after approval, the sponsor intends to roll out the clinical trial to another Member State, this would have to be approved separately.

114.       Under this policy option the involvement of the Commission or the Agency (apart from the single submission point, see above) would be limited to technical support of the joint assessment, and to act as 'facilitator' in the joint assessment.

115.       The impact of this policy option is discussed below (point 5.1.3).

1.1.9. Policy option No 1/4 — Single submission with central assessment by the Agency of issues not related to ethical aspects

116.       As in policy option No 1/3 this option would build on a cooperation of member States in assessing the clinical trial application.

117.       However, this policy option would be modelled after the 'centralised authorisation procedure' for medicinal products. This procedure has been established in 1995 in order to address difficulties in the authorisation of medicinal products. These difficulties are in some respects similar to those for clinical trials today as presented above (2.2.1).

118.       Policy option No 1/4 would consist of a central submission and subsequent central assessment by a scientific committee located and administered within the Agency. The basic principle of the working of the scientific committees located with the Agency is the involvement of all Member States in the Committee structure.

119.       Within the Committee, a 'rapporteur' would be established on the basis of mutual agreement. The 'rapporteur' would be charged with drafting the assessment report.

120.       In case of disagreement of a Member State with the rapporteur's opinion, the Committee would proceed to a vote on the basis of majority voting.

121.       On the basis of the opinion, the Agency (or, if legally required, the Commission) would issue an authorisation decision which would be valid for the entire Union.

122.       In addition, each Member State concerned would issue a national decision covering the ethical aspects of the clinical trial (see point 2.3).

123.       Thus, unlike in policy option No 1/3 (with an 'integrated decision'), this policy option would build on a 'dual decision', as is the case today (see point 2.2), on a "per Member State" basis.

124.       As with the centralised procedure for medicines, the Agency would provide for a secretariat of the responsible EMA Committee. Moreover, each product/procedure would be followed closely by a dedicated team of EMA staff, including a 'team leader' and support staff.

125.       This policy option would be intended to meet the operational objective of reducing administrative burden, as well as other compliance costs. Moreover, it would be intended to meet the operational objective of reducing delays for the start of a clinical trial (see point 3.1). The impact of this policy option is discussed below (point 5.1.3).

1.1.10. Policy option No 1/5 — Choice of legal form — Adopting the text of the Clinical Trials Directive in the form of a Regulation

126.       This is not an alternative to policy options No 1/1 to 1/4, but a possible add-on (cumulative policy option). It focuses on the legal form of the text. This option would replace the Clinical Trials Directive by a Regulation. Unlike a Directive, which only binds Member States as to the result to be achieved while leaving to them the choice of form and methods,[85] a Regulation would obviate the need for national transposition measures.

127.       Adoption of a Regulation would require Member States to repeal, with effect from the date of application of the Regulation, their existing national regulations transposing the Clinical Trials Directive.

1.1.11. Policy option No 1/6 — Combination of policy option No 1/3 and No 1/5

128.       This policy option would 'combine' the policy options No 1/3 (joint assessment) with the policy option No 1/5 (legal form of a Regulation).

1.14. 4.2.    Objective No 2 — Regulatory requirements adapted to practical considerations and needs

1.1.12. Policy option No 2/1 — No action at Union level (baseline option)

129.       This policy option would leave the situation as it is. Member States cannot act or can act in only a very limited manner, as the Clinical Trials Directive is based on the principle of exhaustive harmonisation.

1.1.13. Policy option No 2/2 — Enlarging the scope of non-interventional trials

130.       The Clinical Trials Directive applies only to ‘interventional trials’, but not to ‘non-interventional’ trials.[86] Non-interventional trials are trials which meet all four of the following conditions:

           the medicinal product is prescribed in the usual manner in accordance with the terms of the marketing authorisation;

           the assignment of the patient to a particular therapeutic strategy is not decided in advance by a trial protocol but falls within current practice;

           prescription of the medicine is clearly separated from the decision to include the patient in the study;

           no additional diagnostic or monitoring procedures are applied to the patients and epidemiological methods are used for the analysis of the collected data.

131.       The reason for excluding non-interventional trials from the scope of the Directive is that they typically pose a lower risk than interventional trials. In addition, this restriction is meant to exclude medical activities which are normal clinical practice and, as such, part of the general medical surveillance of a patient.

132.       This policy option would broaden the scope of non-interventional studies by removing the fourth of the four cumulative requirements. This would mean that any study using authorised medicinal products for their authorised indication, even with additional intervention, would fall outside the scope of the Clinical Trials Directive if the subjects are not assigned prospectively, for example by randomisation. Consequently, the administrative burden and other compliance costs would be reduced for these studies which would be in line with the operational objective defined in point 3.2.

133.       This approach was suggested by several stakeholders during the 2011 public consultation.

1.1.14. Policy option No 2/3 — Excluding ‘non-commercial sponsors’

134.       One policy option to address the specific and operational objective is to exclude generally ‘non-commercial sponsors’ from the scope of the Clinical Trials Directive. If ‘non-commercial sponsors’ were excluded, their studies would not be regulated any more by the Clinical Trials Directive and non-commercial sponsors would not be affected by the administrative burdens and other compliance costs caused by this legislation. Thus, the operational objective defined in point 4.1.2. would be achieved.

135.       The approach set out in this policy option applies in the U.S. and in Japan.[87] This policy option was much discussed as an option during the legislative process of the Clinical Trials Directive between 1997 and 2001. In the years following the adoption and entry into force of the Clinical Trials Directive, there were frequent calls for a revision of that text with the aim to excluding 'non-commercial sponsors'.[88]

1.1.15. Policy option No 2/4 — Removing regulatory requirements on the basis of the knowledge of the IMP

136.       As mentioned earlier (see points 2.1.1 and 2.2.2), clinical trials are performed not only with unknown compounds, but also with authorised and well-known medicines. Clinical trials with such authorised/well-known medicines typically pose a low risk compared with normal clinical practice, as the medicine is already on the market and has undergone a marketing authorisation procedure and obtained subsequent approval.

137.       This policy option would remove regulatory requirements for clinical trials with authorised medicinal products used for the authorised indication or with medicines used in a well-known use. This would reduce the regulatory burden and thus costs, thereby contributing to the operational objective defined in point 4.1.2. Moreover, clinical trials with authorised IMPs are typically conducted by non-commercial sponsors. Thus, the operational objective with regard to non-commercial sponsors would be addressed.

1.1.16. Policy option No 2/5 — Insurance/Optional ‘national indemnification mechanism’

138.       Under this policy option Member States would set up a national indemnification mechanism which provides for indemnification for clinical trials performed on their territory, taking account of the national legal system for liability.[89] Such national indemnification mechanism already exists in some of the Nordic Member States.

139.       On the basis of these experiences, such national indemnification mechanism would work as follows:

· The sponsor/investigator, instead of concluding a private liability insurance, has the possibility to refer in the clinical trial application to the national indeminification mechanism. This may be free of change or subject to a payment – depending on the arrangement in the Member State;

· A clinical trial participant suffers a damage in a clinical trial;

· This damage is to be compensated by the sponsor or investigator in accordance with national liability laws (special liability laws or general private law);

· Where the damage is to be compensated, the national indemnification mechanism pays the damages to the subject;

· Depending on the arrangements of the Member State, the national indemnification mechanism turns to the damaging sponsor/investigator to re-coup the compensation payment.

140.       Thus, a national indeminification mechanism would have the following features:

· It would be optional for sponsors to join such an indemnification mechanism. Sponsors who opt out would have to obtain cover on the insurance market;

· Member States would establish the way of financing the national indemnification mechanism. They could either make it subject to a contribution by the sponsor or publicly-funded — at least where the clinical trial is not intended to generate data for a future application for marketing authorisation.

· Apart from these general principles, it would be up to each Member State to decide the details of the national indemnification mechanism and, in particular, whether, in case of a payment to a damaged patient, the national indemnification mechanism can take action against the damaging party (sponsor or investigator).

· The national indemnification mechanism would not interfere with national rules on liability (degree of negligence, if any, burden of proof, etc.).

141.       Such mechanism would greatly facilitate assuring insurance coverage and costs for this coverage would be limited to the costs caused by damage that actually occurs. Administrative burdens and other compliance costs would be reduced (see, for a discussion on the impact, point 5.2.5), thus addressing the operational objective of cutting costs created by the obligatory insurance/indemnity.

142.       Such mechanism could only be established at national level, not at EU-level, as the liability rules in the EU Member States diverge largely in terms of negligence, burden of proof, and compensated damage.

1.1.17. Policy option No 2/6 — Combination of policy option No 2/4 and No 2/5

143.       In this policy option the policy options No 2/4 and No 2/5 would apply cumulatively. This is only relevant insofar as the obligatory insurance/indemnification is concerned: This obligatory insurance/indemnification would not apply for low-risk trials. For other than low-risk trials, the national indemnification mechanism in policy option No 2/5 would apply.

1.15. 4.3.    Objective No 3 — Addressing the global dimension of clinical trials when ensuring compliance with GCP

1.1.18. Policy option No 3/1: Leaving the situation as it is (baseline option)

144.       The ‘self-regulation’ option would mean continuing to rely on:

· Voluntary commitment on the part of sponsors to ensure that clinical trials in non-EU countries are performed in accordance with GCP;

· Regulatory supervision and inspections by non-EU countries in their jurisdictions;

· Some inspections by the inspectors of Member States in the framework of applications for marketing authorisation.

1.1.19. Policy option No 3/2: Facilitating GCP inspections by increasing transparency

145.       This option would put sponsors under an obligation to register publicly all clinical trials whose results are used subsequently in an application for authorisation of a clinical trial or for marketing authorisation for a medicinal product.

146.       Such official public register is already in place in the EU: The 'Clinicaltrialsregister.eu'[90] has been launched in early 2011. This public register is, however, not open for registrations of clinical trials which are performed exclusively in third countries. Rather, this public register is limited to clinical trials which are performed in at least one Member States.[91]

147.       The aim of such public registration would be to allow enforcement authorities to intervene and police these clinical trials. It would also build up pressure for sponsors to comply with GCP.

1.1.20. Policy option No 3/3: Inspections of non-EU countries' regulatory systems for clinical trials

148.       According to pharmaceutical law, any clinical trial in third countries which is referred to in a marketing authorisation application has to be conducted on the basis of principles which are equivalent to those applied in the EU (see point 2.2.3).

149.       Under this policy option the effective application of this 'equivalence rule' would be strengthened by introducing inspections of third country regulatory systems ('system inspection', sometimes referred to as 'audits') in order to verify whether the third country regulatory system, and its control and enforcement, is equivalent to that in the EU as far as subject rights and safety and data robustness is concerned.

150.       Thus, this policy option would put in place a system of inspection of third countries' regulatory systems for clinical trials.

1.1.21. Policy option No 3/4: GCP inspections of non-EU countries' clinical trial sites

151.       This policy option would give the Agency the task of performing inspections in non-EU countries itself, i.e. without drawing on inspection capacity provided voluntarily by Member States. Consequently, the Agency would not have to rely exclusively on inspectors provided by Member States.

152.       Unlike policy option No 3/3 this policy option would not target the regulatory system of the third country, but it would target individual clinical trials sites, sponsor establishments, or establishments of actors to which the sponsor has outsourced certain tasks.

1.1.22. Policy option No 3/5 — Combination of policy option No 3/2 and No 3/3

153.       This policy option would combine a strengthened transparency (policy option No 3/2) with inspections of non-EU countries' regulatory systems for clinical trials.

5.           Impact of policy options

General remarks

154.       In assessing the policy options, the focus is on the social and economic impacts. Regarding environmental impacts, in principle the policy options discussed here do not have a direct or noteworthy indirect impact. However, should there be such an impact this is highlighted in the assessment of the respective policy option. Regarding social impacts, the key aspect to consider is the impact on public health and patient health and safety.

155.       Regarding the economic impacts, it is to be stressed that approximately 9% of the clinical trials are run under the responsibility of an actor which falls within the EU definition of 'Small and Medium Enterprise' ('SME', see Annex 2). This relatively low figure compared to other sectors can be explained by the fact that 'academic sponsors' are not considered as SMEs (see Annex 2 for details). Therefore, all costs created or saved by the policy options concern to approximately 9% SMEs. Where there is a specific impact on SMEs this is going to be specifically highlighted.

156.       Regarding micro-enterprises (see Annex 2), in view of the complexities of the regulatory and business environment for conducting clinical trials in Europe, it can be assumed that there are practically no micro-enterprises active in this sector.

157.       The Commission, in its Communication 'Strategy for the effective implementation of the Charter of Fundamental Rights by the European Union',[92] has committed to examine the impact of legislative proposals on fundamental rights where such an assessment is relevant. The fundamental rights are laid down in the Charter of Fundamental Rights of the European Union ('the Charter').[93] However, in accordance with the Commission guidance in COM(2010)573,[94] this does not mean an examination of the draft act's legal compliance with fundamental rights, which is carried out at a later date on the actual draft act.

158.       The conduct of a clinical trial may impact Article 1 ('Human dignity') and Article 3 of the Charter ('Right of the integrity of the person'). The regulation of the conduct of a clinical trial may impact on Article 13 of the Charter ('Freedom of the Arts and the sciences'), as well as Article 35 of the Charter ('Health care') and Article 16 of the Charter ('Freedom to conduct a business').

159.       The socioeconomic impacts are thus intrinsically linked with impacts on fundamental rights as set out in the Charter:

· Any decrease/increase of patient safety is a negative/positive impact on Articles 1 and 3 of the Charter;

· Any reduction/increase of costs for conducting clinical trials (be they administrative burden or other compliance costs) is to be seen as positive/negative impact on Articles 13, 16 and 35 of the Charter.

160.       Therefore, impacts on fundamental rights are going to be addressed through the assessment of the socioeconomic impacts of each policy option. However, if an impact is particularly critical, this is explicitly highlighted.

161.       Regarding implementation costs (staff and IT) for the Commission and the Agency, it is crucial to assess these in view of the resources available in the Multiannual Financial Framework (MFF) 2014-2020.[95] The impact on implementation costs for any policy option needs to be carefully taken into account.

1.16. 5.1.    Objective No 1 — A modern regulatory framework for submission, assessment and regulatory follow-up of applications for clinical trials

1.1.23. Policy option No 1/1: No action at Union level and reliance on voluntary cooperation of Member States (baseline option)

162.       No action would mean that the current situation would persist. This situation (baseline option) can be described as follows:

1.16.1.1. 5.1.1.1.          Social/health impact

Safety of participants

163.       The Clinical Trials Directive has contributed to ensuring subject safety in clinical trials (see point 2.1). However, lack of coordination and cooperation in the assessment phase and the regulatory follow-up can put subjects at risk: follow-up information generated during the assessment is not shared with other Member States concerned.

164.       Moreover, the ‘patchwork’ of separate assessment procedures for clinical trials by each Member State concerned does not necessarily ensure the highest possible standard of assessment, as the specialist expertise necessary might not always be readily available in every Member State concerned. This works to the detriment of the safety of participants in clinical trials.

165.       Both these points would be addressed by the VHP (see point 4.1.1 for details on the VHP).

Inequalities in access to innovative treatment

166.       The baseline option means that the protocol, conduct and design of the same clinical trial can be subject to different changes and adjustments in the authorisation procedure. These divergences can have an impact on data generated in the trial. In principle, one clinical trial is supposed to be based on one design and to generate one set of data. If the conduct and design of the trial diverge, the integrity of the dataset emerging from it could be compromised.

167.       As a result, the launch of a clinical trial gets delayed and sponsors may even decide to withdraw the clinical trial from one or more Member States (see point 2.2.1.2). This means that patients in those Member States are deprived of the potential benefits of clinical research, which leads to inequalities in public health.

168.       This point would be addressed by the VHP.

1.16.1.2. 5.1.1.2.          Economic impact

1.16.1.3.           Administrative costs/Administrative burdens

169.       At present, the Clinical Trials Directive creates administrative costs of approximately 306 m EUR per year (see Annex 3).

170.       These high administrative costs, despite the relatively low number of clinical trials, are a direct consequence of the regulatory framework set by the Directive (see point 2.2.1.1). It is very labour-intensive and costly to multiply largely identical administrative procedures for multinational clinical trials — and these costs increase even further if requirements differ for individual countries. Sponsors spend a great deal of time retrieving the relevant information, modifying it, and writing the application for authorisation and follow-up information.

171.       These costs are to approximately 80% administrative burdens, as much of this information would not be collected or processed by the sponsor in the absence of legislation. This holds in particular for follow-up information, substantial amendments, annual safety reporting, end of trial reporting, and some aspects of SUSAR reporting (see Annex 3).

172.       Despite development of the VHP, the administrative costs remain largely identical, due to the following factors:

· The VHP does not replace the separate national submission procedures: instead, in fact, it produces ‘two waves’ of submission, one under the VHP and, subsequently, national waves. This holds true both for the initial application and for subsequent substantial amendments;

· The VHP does not extend to other regulatory steps, such as submission of SUSARs, the end-of-trial notification or the annual safety report.

Other compliance costs

173.       Apart from the administrative costs, the Clinical Trials Directive also gives rise to other compliance costs. These add up to approximately 2 200 m EUR per year (see Annex 3).

174.       In terms of other compliance costs, one would expect that the VHP (see point 4.1.1) has a favourable impact as the conclusions on an application for authorising a clinical trial are identical in all Member States participating in the clinical trial. Moreover, this identical outcome does not only hold for the actual authorisation ('yes'/'no'), but also for accompanying conditions and comments. This is critical as those divergencies lead to additional costs (see point 2.2.1.1). Nevertheless, in 2011, of the approximately 1100 multinational clinical trials applied for in the EU, only 84 applications were lodged by sponsors under the VHP. This moderate success rate of the VHP, with only approximately 8% of all multinational clinical trials being submitted through this procedure is due to the following reasons:

· As the VHP is not derived from legislation, but builds on a voluntary initiative of Member States, there is the continuing possibility that, at the end of the VHP process a Member States does change, for its territory, the assessment or conclusions;

· One large Member State refuses participation in the VHP, and several other Member States decide on their participation on a case-by-case basis.

              Delay of launch of a clinical trial

175.       The delays for the full launch of a clinical trial (see point 2.2.1.2) do not only have a health impact (see point 5.1.1.1), but also an economic one: In particular industry sponsors have a strong interest in launching a clinical trial quickly once the protocol is finalised. Any unnecessary lengthening of the development process, and be it just 1-2 months, has to be avoided in order to justify investments, and in order to be able to reach quickly the marketing stage.

176.       While the VHP shortens the delays which are due to discrepant assessments, it also adds delays: The 'two waves' of submissions (see above under 'administrative costs') lead to an additional delay before a clinical trial is authorised: Once the (non-legally binding) conclusions in the VHP have been made, the actual, formal request for authorisation has to be submitted again to each Member State which leads to an additional approval timeline.

Implementation costs[96]

Resources in Member States

177.       Member States have approximately 112 FTEs available in NCAs (see Annex 2) who work specifically on the assessment and follow-up (except safety reporting) of clinical trials (including validation staff, excluding administrative support staff, external resources, and inspection personnel). The personnel in the ECs must be added to this. It is not possible to give a figure for FTEs in ECs, as:

· EC members are usually not full-time members; and

· ECs also assess other research in addition to clinical trials.

178.       There are approximately 950 ECs in the EU entitled to issue a ‘single opinion’ (see Annex 2). Each EC has approximately 6 to 15 members.

Resources in the Agency and in the Commission

179.       The role of the Agency in application of the Clinical Trials Directive is limited to administering EudraCT and to coordinating GCP inspection activity in the centralised authorisation procedure for medicines. For these tasks, the Agency has approximately 2 FTEs.

180.       In the Commission, 0.25 FTEs are assigned to all aspects of regulation of clinical trials. This is in line with the financial statement attached to the 1997 proposal for the Clinical Trials Directive.[97]

1.16.1.4. 5.1.1.3.          Further development in the absence of EU action

181.       In the absence of action at EU level, the situation as set out in this baseline scenario would not improve. Rather, in view of the developments which are expected in terms of research with pharmaceuticals (see point 2.2.1.4), if no action is taken the situation is going to aggravate further in terms of social/health impact and economic impact. In particular the VHP does not sufficiently address the problems set out in point 2.2.1: besides the shortcomings highlighted in point 5.1.1.2, the VHP does not address the issue of multiple submissions (which is an important driver of administrative costs) and does not sufficiently address the risk of diverging assessments, leading to additional costs and to the delay of the launch of a clinical trial.

182.       Therefore, a careful projection of the current situation (set out in point 2.2) into the future has to lead to the conclusion that, if no action at EU level is taken now to reach the objectives set out in section 3.1, the situation is going to aggravate further both in terms of public health and in terms of costs.

1.1.24. Policy option No 1/2 — Single submission with separate assessment

1.16.1.5. 5.1.2.1.          Social/health impact

183.       As regards the impact on terms of health and patient safety there would be no change compared with the present situation. In particular:

           In terms of social/health impact, there would be no change compared with the baseline option. In particular, this policy option would not bring gains in patient protection: the level of protection would depend on the (differing) assessments by the Member States;

           The separate assessments would lead to differing conclusions as regards the protocol and, thus, to differing versions of the protocol for the same clinical trial.

1.16.1.6. 5.1.2.2.          Economic impact

184.       This policy option would reduce administrative costs to 45.5 m EUR, i.e. it would save administrative costs of 260.5 m EUR per year compared with the baseline option. Moreover, the share of administrative burdens would decrease more than proportionately, as many multiple reporting obligations would become obsolete in this policy option (see Annex 3). As the assessment procedure would be identical to policy option No 1/1, the administrative costs for follow-up information would remain identical. This is also recognised by stakeholders. During both public consultations all types of stakeholders (patients, non-commercial and industry sponsors) welcomed explicitly the idea of a single submission point, while stressing that the issue of diverging assessments would need to be addressed, too. While many questions on operational details were raised (e.g. related to confidentiality, archiving, authentification, and personal data protection), the policy option in itself was hailed as 'the only way forward'[98] which would 'greatly reduce the administrative work of sponsors'.[99] Also Member States 'endorsed' this policy option highlighting that 'it may be helpful for sponsors, reduce administrative burden and might facilitate the conduct of clinical trials in EU […]'.[100]

185.       In terms of other compliance costs, however, the situation would be identical to policy option No 1/1, as this policy option is limited to an IT-tool to submit information.

186.       In terms of implementation costs, the one-off costs for IT and to running costs vary depending on the technical solution (see Annex 6).

           'Extensive IT solution' (suggested by the Agency)[101]: One-off costs would be 6.3m EUR. Running costs would be 1.26m EUR per year. To this add 19 FTEs (11 Administrators and 8 Assistants);

           "Limited IT solution" (suggested by the Commission)[102]: One-off costs would be 1.62m EUR. Running costs would be 0.34m EUR per year. In addition, 0.25 FTEs are required to provide regulatory expertise.

187.       The choice as to which solution is to be pursued is intrinsically linked to the decision as to where the single submission point is located: at the Agency or at the Commission (see Annex 6, point 2 - 'financing strategies'). This would have to be a political decision. A detailed list of arguments to support this decision-making is contained in Annex 8.

1.16.1.7. 5.1.2.3.          Other aspects

188.       This policy option would greatly simplify the regulatory framework for the authorisation and regulatory follow-up of clinical trials. The multiple submissions would be replaced by a 'one stop shop'.

1.1.25. Policy option No 1/3 — Single submission with joint assessment by Member States of issues not related to ethical aspects

1.16.1.8. 5.1.3.1.          Social/health impact

189.       Protection and the safety and rights of participants would improve, as compared with the baseline option, as expertise of different Member States would be brought together: This policy option would ensure that the Member States concerned cooperate on the non-ethical aspects of approval of clinical trials. Such joint exercises could spot any flaws in the assessment and hitherto undetected risks, thus improving the protection given to the subjects and the quality of the clinical research. Moreover, access to clinical trials would be facilitated: As the assessment and conclusions for a clinical trial would be identical situations would be avoided where a clinical trial is not performed in a given Member State due to incompatible requests for changes to the protocol (see point 5.1.1.1).

190.       In terms of delays, this policy option would only involve the Member State concerned, i.e. it would involve in practice rarely more than 6-8 Member States (see point 2.1.1). Experience in the VHP has shown that it is in practice well possible for the Member States concerned to agree on the assessment of a clinical trial application within 60 days. Therefore, it can be expected that the deadlines which exist today for approval of a clinical trial (60 days, see point 2.1.2) is maintained. Moreover, a joint assessment by Member States would remove delays for the start of a clinical trial which occur if the protocol and trial design has to accommodate conflicting assessments and request from different Member States (see point 2.2).

1.16.1.9. 5.1.3.2.          Economic impact

191.       In terms of administrative costs, this policy option would have largely the same impact as No 1/2. In addition, however, there would be a further reduction linked to follow-up submissions, which are administrative burdens (see Annex 3). In this policy option, as the authorisation dossier submitted would be assessed jointly by the Member States concerned, there would also be joint submission of follow-up information.

192.       This policy option would reduce administrative costs to 34.3 m EUR, a saving of 271.7 m EUR per year compared with the baseline option (see Annex 3).

193.       These savings all concern administrative burdens, i.e. these costs are not going to incur if legislation did not impose them (a sponsor would not voluntarily submit an application file to a Member State individually, if this is already done through a single submission point).

194.       In terms of other compliance costs, this policy option would ensure that the conclusions on an application for authorising a clinical trial are identical in all Member States participating in the clinical trial.[103] This would not only hold for the actual authorisation ('yes'/'no'), but also for accompanying conditions and comments. This would ensure that the same protocol applies in each Member State where the trial is intended to be performed. While it is not possible to quantify these savings to the same degree of precision as for administrative costs, the estimated saving would be in the range of 440 m EUR per year (see Annex 3), i.e. other compliance costs of 1 760 m EUR.

195.       It is in particular with a view to the impact on administrative and other compliance costs, as well as in a view of delays, that the majority of all stakeholder groups supported this policy option. For example, practically all non-commercial sponsors welcomed that this policy option "provides a crucial opportunity to implement a 'risk-based' approach that is consistent across Member States."[104] All but one[105] patient associations who responded to the 2011 public consultation favoured this policy option.[106] Pharmaceutical companies and associations – with the exception of some of the very large pharmaceutical companies and EFPIA[107] – supported this policy option, stressing that the assessment of the clinical trial in accordance with this policy option was "the assessment of choice."[108] Finally, Member States favoured this policy option[109] stressing that some elements of this policy option "could profit from the VHP experience" and expressing "strong support" for the "principle of keeping the clinical trial approval at the national level maintaining the Member State responsibilities on clinical trials conducted in their territories".[110]

196.       In terms of implementation costs, this policy option would reduce costs on the part of the NCAs to the extent that the assessment would be performed in greater depth by just one Member State. However, the basic principle of this policy option is an assessment by all Member States concerned. Therefore, these savings, if any, would be minor.

197.       Under this policy option the involvement of the Commission or the Agency (apart from the single submission point, see above) would be limited to technical support of the joint assessment, and to acting as 'facilitator' in the joint assessment (see point 4.1.3). Thus, the role of the Commission or the Agency would not include follow-up of individual authorisation procedures, such as contacts with the applicant, or (assisting in) drafting assessments or grounds for non-acceptance.

198.       Rather, the role of the Commission or Agency would be limited to the following:

           Providing meeting room capacities for meetings, where necessary;

           Preparing and chairing meetings of Member States in order to ensure coherence of the general functioning of the joint assessment procedure with procedural requirements set out in the legislation, including respect of timelines.

199.       As set out in Annex 7, this role can have a varying degree of resource needs, depending on whether an 'extensive support structure' (suggested by the Agency) or a 'limited support structure' (suggested by the Commission) would be chosen:

           'Extensive support structure'[111]: additional resource needs compared to the baseline option would be 7 FTEs (3 administrators and 4 assistants), plus 48.5% overhead.

           'Limited support structure'[112]: additional resource needs compared to the baseline option would be 1.5 FTE (all administrators, including overhead).

200.       Apart from staff, there are travelling reimbursement costs at EU level which range between 102 000 EUR and 210 600 EUR per year, depending as to whether the support is provided by the Commission (i.e. meetings take place is Brussels) or the Agency (i.e. meetings take place is London). For details, see Annex 7.

201.       The choice as to the scale of the support structure is linked to the decision as to who provides the support structure: the Agency or at the Commission (see Annex 7, point 2 - 'financing strategies'). This would have to be a political decision. A detailed list of arguments to support this decision-making is contained in Annex 8.

1.16.1.10. 5.1.3.3.            Other aspects

202.       In this policy option the assessment would be performed only by those Member States where the clinical trial is to take place. Member States where the clinical trial is not intended to be rolled out would not take part in the procedure. This raises the question of how to deal with any subsequent roll-out of a clinical trial (‘staggered launch’). In practice, a clinical trial sometimes has to be rolled out in more Member States than originally planned — for example, in order to meet subject recruitment targets.

203.       This policy option would therefore have to include a mechanism allowing regulatory approval in additional Member States who join the trial after the initial approval. Even if this additional roll-out happens quickly, it could lead to additional delays and costs for staggered launches of clinical trials.

1.1.26. Policy option No 1/4: Single submission with central assessment by the Agency of issues not related to ethics

1.16.1.11. 5.1.4.1.            Social/health impact

204.       In terms of social/health impact, this policy option has the benefit of involving all Member States, thus assembling the best expertise of regulators available to the administrations in Europe.

205.       However, this option might lead to additional delays in authorisation of clinical trials for the following reasons:

· The system of 'dual decision' (national and EU levels, see point 4.1.4) is likely to lead to contradictions. For example, while, at EU level, a specific condition for a clinical trial might be introduced, this very condition may not be compatible with the requirements set out at national level. These contradictions may lead to the delays which are already currently being experienced (see point 5.1.1.1);

· The principle functioning of EMA committees, which is based on the principle of involvement of all Member States in the scientific committees, leads to the involvement of Member States which are not necessarily concerned. This would increase the complexity of the discussions, which takes time;

· A ‘committee structure’ would take the flexibility out of the authorisation procedure. Today, the authorisation process in the Member States is highly flexible, with recurrent, and also informal, contacts between the assessors and the sponsor. A heavy, very formal, committee structure would deprive sponsors of these advantages;

· Management by a committee which would meet only occasionally (e.g. once a month) would lead to further delays.

206.       In view of some of these arguments it is not realistic to assume that sponsors would receive approval or non-approval of a clinical trial within the current timelines of 60 days. Rather, it has to be expected that a minimum of 90 days is required to assess the application dossier and to reconcile the view of all Member States.

207.       In addition, this policy option would lead to an ‘institutional connection’ and ‘continuum’ between the authorisation procedure for clinical trials throughout development of a medicinal product and the marketing authorisation of the resultant product.[113] It could be argued that this is very positive, as the body in charge of assessing new medicines (i.e. the Agency) is also involved in steering the clinical research studies. However, it is crucial to bear in mind that authorisation of a clinical trial must not be confused with authorisation of a medicinal product: in the former, the assessment looks at the benefits and risks of a treatment for a patient (be it a patient in the experimental arm of the trial or in its control arm). This is done in the absence of certainty or knowledge about which is the most favourable treatment. Applications for marketing authorisation, on the other hand, are assessed on the basis of the medicinal product and its intrinsic properties (see point 2.1.2).

208.       In other words, for a marketing authorisation it might be desirable to have data from a specific clinical trial, whereas, from the viewpoint of regulation of clinical trials, this trial should not be approved, considering the benefits and risks to the patient.

209.       Although this seems paradoxical, this is a logical consequence of the ethical limits to performing clinical research on humans. The basic principle of these limits is that ‘in medical research involving human subjects, the well-being of the individual research subject must take precedence over all other interests’.[114]

1.16.1.12. 5.1.4.2.            Economic impact

Administrative costs/administrative burden

210.       In terms of administrative costs, this policy option would lead to dual approval and, thus, two sets of follow-up questions. This would lead to administrative costs of 41.8 m EUR, i.e. savings of 264.2 m EUR. The share of administrative burdens would be similar to the option No 1/1.

211.       In terms of other compliance costs, the impact would be similar to policy option No 1/3, i.e. savings of approximately 440 m EUR.

212.       In terms of implementation costs, these would relate largely to an additional role of the Agency. As set out in Annex 3, it can be estimated that the additional staff needs would be in the range of 4 000 FTEs per year.

213.       In view of these implementation costs, it is crucial to be aware that they would have to be covered through specific fees. These fees would be collected by the Agency, too. In view of the number of clinical trials, these fees for authorisation of a clinical trial would be substantial: This would impact in particular on SMEs and academic sponsors who do not necessarily have the financial resources to cover regulatory fees.

214.       The increase in implementation costs, which would be passed on to sponsors in the form of fees, were also the key concern of SMEs and academic sponsors the responses to the two public consultations. Research networks, for example acknowledged that the centralised marketing authorisation procedure works well and has been a success, but stressed that this model would be 'unworkable' for clinical trials.[115] Non-commercial sponsors acknowledged that this policy option was "appealing, but […] that a procedure involving all Member States in all aspects of each application would cause too much administration. If this […] would lead to an increased fee, it would definitely be a problem for academic researchers in general."[116] Equal concerns were raised by industry associations representing smaller pharmaceutical companies which stressed the need to maintain flexibility, and recalled that very few clinical trials are conducted in more than 5-6 Member States.[117] The very large pharmaceutical companies, as well as EFPIA, however, did consider costs and complexities of this policy option as surmountable. Their responses, however, base their opinion on the idea of a 'pool of appropriate experts drawn from across the Member States'[118] which would act in full independence from national affiliation. Thus, these responses do not acknowledge the basic principle of inclusion of experts from all Member States into scientific committees at the EMA. The idea of limiting the involvement of the assessment of a clinical trial application to a 'pool of experts' is unrealistic in view of the sensitivity of the matter discussed, i.e. the potential exposure of humans to a medical experiment.

215.       For the reasons set out above all Member States strongly opposed this policy option in the 2011 public consultation. Patient organisations viewed this policy option in comparison with policy option No 1/3 and expressed, with one exception,[119] support for the latter.

1.16.1.13. 5.1.4.3.            Other aspects

216.       Voting in case of disagreement: Apart from the delays it may cause, the '(qualified) majority vote' in case of disagreement, as provided in this policy option, raises doubts in terms acceptability and feasibility for the following reasons:

           It would mean that subjects in a given Member States would be exposed to a clinical trial (i.e. a clinical experiment) without the consent of that Member State who is in charge of supervision the conduct of the trial;

           In practice, any result of a majority vote would be circumvented by the outvoted Member State by arguing that the matter at stake touches on ethical issues.

217.       Additional roll-out: This policy option would, to some extent, facilitate roll-out to additional Member States (see point 5.1.3, 'other aspects'). However, any such additional roll-out would not be automatic. Instead, it would require an additional assessment of the ethical aspects for each Member State.

1.1.27. Policy option No 1/5 — Choice of legal form — Adopting the text of the Clinical Trials Directive in the form of a Regulation

218.       This policy option would ensure that the Member States would base their assessment of an application for approval of a clinical trial on an identical text, rather than on diverging national transposition measures.

219.       Moreover, the legal form of a Regulation would provide a more detailed, binding manner to address the procedure for submission of applications for authorisation and for notification of substantial amendments.

220.       In practice, experience shows that transposition of the Clinical Trials Directive has been incorrect and has often given rise to additional procedural requirements. This difficulty would be removed with this policy option.

221.       Moreover, this policy option would have an important simplification effect. The replacing of transposition measures at national level allows the relevant actors to plan and conduct the clinical trial, including multi-national clinical trials, on the basis of one regulatory framework, rather than on the basis of a 'patchwork' of 27 national frameworks in the transposing Member States laws.

222.       However, this policy option does not address diverging interpretations and implementing practices. This was stressed in particular by Member States during the 2009/10 public consultation who argued that, even if the legal form was a Regulation, requirements would still be interpreted differently by Member States bodies in the practical application, unless a cooperation mechanism is in place.[120] Therefore, while the legal form of a Regulation would help to achieve the objective, it is not a solution on its own. It would only contribute in conjunction with one of policy options No 1/3 or 1/4 (see above).

223.       This policy option was presented to stakeholders in the 2009/10 public consultation. Practically all industry sponsors and a large part of the non-commercial community "undoubtedly preferred"[121] the legal form of a Regulation, highlighting in particular the simplification effect and the difficulties for Member States to cooperate if each Member States works on the basis of 'similar, but different'[122] national transposing laws. Amongst non-commercial sponsors, however, there were also voices favouring a Directive as this legal form "would leave more room for interpretation for practical use."[123] Regarding the question whether the legal form of a Regulation would increase or lower the substantial requirements, fears were voiced that these requirements may increase[124] or decrease.[125] In response to these concerns it has to be stressed that the substantial requirements set out in a Regulation would, as in a Directive, be guided by the principles of proportionality and appropriateness while taking account the Treaty on the Functioning of the European Union whereby "The Commission, in its proposals […] concerning health, safety, environmental protection and consumer protection, will take as a base a high level of protection taking account in particular of any new development based on scientific facts"[126] and whereby "A high level of human health protection shall be ensured in the definition and implementation of all Union policies and activities."[127]

1.1.28. Policy option No 1/6 — Combination of policy option No 1/3 and No 1/5

224.       This policy option would 'combine' the policy options No 1/3 (joint assessment) with the policy option No 1/5 (legal form of a Regulation).

225.       It would strengthen policy option No 1/3, as the cooperation amongst Member States in policy option No 1/3 would be facilitated if this cooperative work was based on identical legal provisions.

1.1.29. Comparison of policy options for objective No 1 and synergies

226.       It has to be stressed that the baseline situation is insufficient to address the problem. This was highlighted repeatedly during the two public consultations by all stakeholders (research community, industry and patients) and also by Member States. Indeed, the launch of the VHP by Member States, without any legal basis and as a purely voluntary initiative, is a sign that the baseline option is unsatisfactory.

227.       While policy options No 1/2, No 1/3 and No1/4 have one common element (the single submission point), they are mutually exclusive.

228.       The common element, which is part of policy options No 1/2, No 1/3 and No 1/4 greatly reduces administrative costs and burdens and thus contributes to addressing the problem.

229.       Policy option No 1/2, however does insufficiently address issues of separate assessments of identical issues in relation to the same clinical trial. In this respect, policy options No 1/3 and 1/4, which address not only the submission process, but also the assessment process of a clinical trial application, are to be favoured.

230.       Policy option No 1/3 is also superior to the baseline option with the VHP: There are various structural shortcomings of the VHP which cannot be remedied in the baseline option, but which are addressed with the policy option No 1/3. In particular:

           Policy option No 1/3 provides, unlike the VHP, for a structured cooperation mechanism with legally-binding and enforceable timelines for the cooperation of the Member States;

           Policy option No 1/3 has, unlike the VHP, a clear scope of the joint assessment: Thus, Member State cannot 'escape' a coordination of the assessment by claiming that a given issue is of an intrinsically national or ethical nature;

           By issuing one 'integrated decision' (see point 4.1.3), the procedure in policy option No 1/3 ensures that the assessment of intrinsically national or ethical issues does not run counter the agreement found between Member States in the joint assessment. The VHP in the baseline option does not address this issue.

           Policy option No 1/3 provides for strict requirements for the 'opt-out' of a Member State from the joint assessment of the application for the conduct of a clinical trial. This makes the outcome of the authorisation process more reliable and predictable than under the VHP in the baseline option.

           Unlike in the VHP of the baseline option, policy-option No 1/3 does not require additional submission of a request for authorisation once the joint assessment of the Member States has been finalised.

           Policy option No 1/3 ensures, unlike the VHP in the baselin-option, that all Member States have to participate in the joint assessment of a clinical trial application, and that this is not left to a case-by-case decision of the Member State concerned.

231.       In view of these aspects, policy option No 1/3 is considerable more effective than the VHP of the baseline option.

232.       When comparing policy options No 1/3 and No 1/4, it has to be borne in mind that policy option No 1/4 sets up a very heavy system. It involves every Member State, which is not necessary in view of the roll-out of clinical trials. For example, in 2010, of the 4 400 clinical trials applied for in the EU, only 168 (approximately 4 %) were to be rolled out in eight Member States or more (see Annex 2). Considering this, in view of the additional delays, it seems disproportionate to involve every Member State through a committee structure in the assessment of a clinical trial application.

233.       This holds even more as the operational objectives as regards costs and delays are achieved, in policy option No 1/3, equally well as in policy option No 1/4: The high implementation costs of policy option No 1/4, with a considerable increase of personnel at EU-level, do not justify the benefit of a streamlined authorisation procedure, if the same effect can be achieved with the resources required in policy option No 1/3.

234.       Moreover, policy option No 1/4, with its rather heavy procedure, would be of little interest to academic sponsors, who typically run clinical trials in fewer Member States than the pharmaceutical industry’s very large trials during late stages of product development. In addition, potential fees would create difficulties for SMEs who have limited financial resources for regulatory purposes.

235.       Added to this, policy option No 1/4 adds new complexities to the approval procedure, which would be avoided in policy option No 1/3. These stem from the dual approval in policy option No 1/4, which would be necessary in this policy option to take intrinsically national and ethical issues into account. Such dual approval would be avoided in No 1/3.

236.       Policy option No 1/3, on the other hand, provides a ‘slimmer’ procedure. For the initial authorisation, it involves only the Member States where the clinical trial is to be performed (a mechanism would have to be set up to allow roll-out to additional Member States subsequently). Under policy option No 1/3 approval is also likely to be cheaper and faster than in No 1/4. This is in particular of interest for academic research and SMEs.

237.       Policy option No 1/5 has the benefit of addressing divergent approaches in Member States which do not stem from the application of EU-rules, but from their transposition into national law.

238.       Policy option No 1/6 is identical to policy option No 1/3, with policy option No 1/5 as add-on. It would help to ensure a coordinated approach in assessment of a clinical trial and follow-up action, based on identical criteria. This can only really be ensured if the EU legislation is not transposed into 27 separate national laws. While it is very difficult to quantify this impact, from a qualitative viewpoint it is highly relevant.

              Overview — Impact of policy options to address objective No 1

|| Contribution to addressing the problem[128] || Health/social impact compared with baseline[129] || Economic impact/costs (in EUR) for sponsors compared to baseline || Implementation resources/costs for EMA/COM || Other comments

Other compliance costs || Administrative costs || Costs (other than resources) || Resources

Policy option No 1/1 (baseline) || (o) || (o) || 0 || 0 || - || - ||

Policy option No 1/2 (single submission with separate assessment) || (+) || (=) || 0 || - 260.5 m || One-off: between 1.62m (limited IT solution) and 6.3m (extensive IT solution) Running: between 0.34m (limited IT solution) and 1.26m (extensive IT solution) || Running: between 0.25 FTEs (limited IT solution) and 19 FTEs (extensive IT solution) ||

Policy option No 1/3 (single submission with joint MS assessment) || (+++) || (+++) || - 440 m || - 271.7 m || As in policy option No 1/2 || As in policy option No 1/2 Additional 1.5 or 7 FTEs (depending on choice for limited or extensive support structure) ||

Policy option No 1/4 (single submission with central assessment by the Agency) || (+++) || (++) || - 440 m || - 264.2 m || As in policy option No 1/2. || 4 000 FTEs || Longer delays for approval. 'Continuum' between clinical trials authorisation and medicines authorisation.

Policy option No 1/5 (Regulation vs. Directive) || (++) || (+++) || - || - || - || - || Add-on to options No 1/2 to 1/4

Policy option No 1/6 (Combination of Policy option No 1/3 and 1/5) || (+++) || (+++) || - 440 m || - 271.7 m || As in policy option No 1/2 || As in policy option No 1/2 Additional 1.5 or 7 FTEs (depending on choice for limited or extensive support structure) ||

1.17. 5.2.    Objective No 2 — Regulatory requirements adapted to practical considerations and needs

239.       The policy options discussed in this chapter for achieving objective No 2 directly impact on the two regulatory requirements which were highlighted by stakeholders in both public consultations as particularly disproportionate and burdensome: the obligatory insurance/indemnity and the annual safety report.

240.       The impact of the individual policy options on these two regulatory requirements is discussed below.

1.1.30. Policy option No 2/1: No action at Union level (baseline option)

1.17.1.1. 5.2.1.1.          Social/health impact

Obligatory insurance/indemnity

241.       The obligatory insurance/indemnity ensures that, in case of damages caused by a clinical trial, the subject receives compensation — irrespective of the financial means of the sponsor or investigator. This helps to protect clinical trial subjects.

Annual safety report

242.       The annual safety report can be a useful tool for NCAs or ECs to supervise and follow up the safety profile of an IMP, particularly if the compound is still largely unknown and not yet authorised.

1.17.1.2. 5.2.1.2.          Economic impact/costs

Obligatory insurance/indemnity

243.       The yearly costs for obligatory insurance/indemnity for ongoing clinical trials in the EU are approximately 75 m EUR, plus administrative costs of 1.9 m EUR (see Annex 4). On the other hand, approximately 0.025 % of all subjects successfully claim compensation for damages suffered in a clinical trial. Each damages claim is worth, on average, between 3 000 and 6 000 EUR (see Annex 4).

Annual safety report

244.       The costs for drawing up and submitting the annual safety report are approximately 147.8 m EUR per year, to which administrative costs of 5.3 m EUR must be added (see Annex 4).

1.1.31. Policy option No 2/2 — Enlarging the scope of non-interventional trials

245.       The impact of this policy option would be limited to phase IV studies as, by definition, only phase IV studies concern authorised IMPs used in the authorised indication.

246.       If the scope of non-interventional trials were broadened in line with the definition set out above (see point 4.2.2), this would exclude approximately 50 % of phase IV studies from the scope of the EU regulation of clinical trials.

1.17.1.3. 5.2.2.1.          Social/health impact

247.       The immediate impact would be that these studies would be regulated at national level by Member States. Depending on the measures taken by each Member State, this would mean tighter, looser or no regulation of this type of study.

248.       However, this would also undermine past and future efforts to harmonise these studies and would introduce differences in protection of trial subjects and robustness of clinical data generated in the EU. Moreover, in the medium term this would make it more cumbersome to conduct these studies in the EU.

249.       It was against this background that the majority of all stakeholder groups opposed this policy option, but rather supported a wide definition with a risk-based approach (see point 4.2.4).[130] Some sponsors (both industry and 'non-commercial' sponsors) supported this policy option. However, these respondents called, at the same time for a separate regulatory regime at EU level for non-interventional studies.[131] Member States opposed this policy option.[132]

1.17.1.4. 5.2.2.2.          Economic impact/costs

250.       As set out in Annex 4, this policy option would generate the following savings:

Obligatory insurance/indemnity: 3.92 m EUR other compliance costs, plus 123 480 EUR administrative costs.

Annual safety report: 13.06 m EUR other compliance costs, plus 95 445 EUR administrative costs.

251.       However, depending on the measures taken by each Member State, these costs could be pushed up again by regulatory action at Member State level.

1.1.32. Policy option No 2/3 — Excluding ‘non-commercial  sponsors’

252.       In 2010, some 1 620 clinical trials by ‘non-commercial sponsors’ were authorised. These involved 2 037 applications and 93 242 patients (see Annex 2).

1.17.1.5. 5.2.3.1.          Social/health impact

253.       In terms of impact, this would mean that subjects enrolled in a clinical trial run by a ‘non-commercial sponsor’ would not be protected at EU level. Nor would the EU rules ensuring the robustness and reliability of data apply.

254.       This would be a major drawback in terms of a creating a level playing field for conducting clinical trials in the EU without compromising on protection of rights and safety of patients in the EU and data robustness.

255.       This policy option would also have a negative impact on public health in general. Clinical trials run by ‘non-commercial sponsors’ can have a crucial impact on public health as the results may be published and, thus, impact to the choice of treatment options and treatment in general. Publication could also trigger further research into, for example, extension of indications of medicinal products or reduction of the use dosage.

256.       Moreover, if clinical trials by ‘non-commercial sponsors’ were excluded from the scope of the Clinical Trials Directive they would not be subject to harmonised rules at EU level. Member States would again be responsible for regulating these trials via national laws. This would introduce differences in protection of trial subjects in the EU. This, in turn, would make conducting these studies in the EU more cumbersome, which is not in the interest of ‘non-commercial sponsors’ performing clinical trials in different Member States.

257.       These were also the main arguments put forward by stakeholders in the public consultations of 2009/10 and 2011. Indeed, in both public consultations there was unanimity that this policy options should not be pursued. This is a remarkable development over the past 10 years as, during the legislative discussions on the Clinical Trials Directive and in the years thereafter there were frequent calls for excluding non-commercial sponsors from the scope of the Directive altogether (see point 4.2.3).

1.17.1.6. 5.2.3.2.          Economic impact/costs

258.       As set out in Annex 4, this policy option would generate the following savings:

              Obligatory insurance/indemnity: 14 m EUR other compliance costs, plus 488 880 EUR administrative costs.

Annual safety report: 59.9 m EUR other compliance costs, plus 437 400 EUR administrative costs.

259.       Depending on the measures taken by each Member State, these costs could be pushed up again by regulatory action at Member State level.

1.1.33. Policy option No 2/4: Removing regulatory requirements on the basis of the knowledge of the IMP

260.       Under this policy option, the requirements for obligatory insurance/indemnity and the annual safety report would be removed for clinical trials where the IMP is sufficiently known, i.e. authorised, and used within the authorised indication.

1.17.1.7. 5.2.4.1.          Social/health impact

261.       Clinical trials with authorised medicinal products pose a risk to public health which is only minimally higher to that posed by standard care, if at all. This is because the IMP in the clinical trial has already undergone an authorisation procedure. Its safety profile is therefore sufficiently known.

Obligatory insurance/indemnity

262.       Removing the obligatory insurance/indemnity would have no discernible impact on subject protection. Annex 4 shows that the likelihood of an event causing damage is minimal. Based on the figures available, approximately 0.025 % of all subjects enrolled in a clinical trial can be expected to suffer damages which qualify for compensation. While there is no reliable data on this aspect, it is very likely that these damages occur in the setting of non-authorised medicinal products. Moreover, a number of additional types of insurance cover treatment with an authorised medicine, such as:

· Product liability insurance of the marketing authorisation holder for the authorised medicine;

· Professional negligence insurance of the treating physician; and

· Liability insurance of the hospital or healthcare institution where the subject is being treated.

263.       In practice, one of these policies, rather than the insurance/indemnity for damages suffered in a clinical trial, is likely to cover any damages.

264.       In the 2011 public consultation, where this policy option was explicitly put forward, voices diverged. The views of sponsors as well as patient's associations (who are ultimately the beneficiary of the insurance) were divided: While it was highlighted that 'lifesaving treatments cannot be abandoned simply because of the high cost of the insurance,'[133] it was also stressed that risks change and are not always full known. Member States were largely opposed to this policy option as regards obligatory insurance/indemnity.[134] The two national insurer's associations who responded to the public consultation were opposed.

Annual safety report

265.       The absence of an annual safety report for this clinical trial would also have no impact on subject safety. This is because, irrespective of the clinical trial, under the EU legislation on medicinal products[135] each authorised medicinal product is subject to a ‘periodic safety update report’ (PSUR), to be drawn up by the marketing authorisation holder. The PSUR is a very useful instrument to assess the safety profile of a compound, as it is based on the broad data on daily use of the medicine and is drawn up by the marketing authorisation holder who might have a better understanding of the compound than a sponsor.

266.       This aspect were also the main reason for the clear support for this policy option in both public consultations by all stakeholder groups where in particular issues of safety reporting were raised.[136]

1.17.1.8. 5.2.4.2.          Economic impact/costs

267.       As set out in Annex 4, this policy option would generate the following savings:

Obligatory insurance/indemnity: 7.84 m EUR other compliance costs, plus 246 900 EUR administrative costs.

Annual safety report: 26.1 m EUR other compliance costs, plus 190 890 EUR administrative costs.

268.       The savings as regards administrative costs all concern administrative burdens, i.e. these costs are not going to incur if legislation did not impose them: A sponsor would not cover a clinical trial with an authorised medicine in an insurance scheme, as other insurances (product liability insurance, professional negligence insurance of the physician and the hospital, etc. See point 4.2.4) are available. Neither would a sponsor submit an annual safety report unless this was legally required.

1.1.34. Policy option No 2/5 —Insurance/optional ‘national indemnification mechanism’

1.17.1.9. 5.2.5.1.          Social/health impact

269.       A national indemnification mechanism along the lines set out in point 4.2.5 would give the same assurance of compensation for any subject suffering damages as the obligatory insurance/indemnity currently required by the Clinical Trials Directive.

270.       Indeed, some Nordic Member States (Denmark and Finland) already have a system like that proposed in this policy option in place.

1.17.1.10. 5.2.5.2.            Economic impact/costs

271.       In terms of costs, the following differentiation has to be made.

           Costs for covering damages (other compliance costs): While — as with any medical intervention — the potential damage might be high[137], the actual damage caused by clinical trials is very low. As set out in Annex 4, it can be assumed that the successful claims for damages could total approximately 135 000 EUR per year.

           Costs for Member States to run the national indemnification mechanism (implementation costs): The implementation costs for Member States are relatively low. As indicated in Annex 4, dealing with all incoming claims for damages caused by clinical trials would require approximately 9.75 FTE staff for the entire EU, which is equal to approximately 682 500 EUR per year (see Annex 4).

272.       Depending on how the national indemnification mechanism is financed, these two types of costs may be passed on, in a second-round effect, to the damaging party (sponsor or investigator). In this case the contributions of the sponsor would have to cover costs of 817 500 EUR per year (see Annex 4). Compared to the baseline (75m EUR other compliance costs), this is a reduction of other compliance costs of 74.18m EUR.

273.       In addition, there are the administrative costs: These would be limited to a bare minimum, as confirmation of insurance/indemnity cover by a Member State should suffice to comply with the regulatory requirement of insurance. The administrative costs of this policy option can be estimated at approximately 238 000 EUR. Compared to the baseline (1.9m EUR), this is a reduction of administrative costs of 1.66m EUR.

274.       In the 2011 public consultation this policy option was put forward for comments by stakeholders. The responses varied: Some respondents, in particular from the group of non-commercial sponsors welcomed the concept, stressing that "the necessary funds required will be much less than what would be spent by public funding agencies on insurance costs."[138] Other responses from sponsors criticised this policy option and pointed at the risks of increased bureaucracy, and the risk of divesting liability to the state. Most stakeholders from all stakeholder groups had additional questions and concerns in relation to whether such mechanism would be funded by fees or otherwise, and whether such mechanism would be able to turn to the damaging party to re-cover compensation payments to patients. In response to these concerns it has to be stressed that this policy option would leave these matters to the Member State setting up this mechanism (see point 4.2.5). Only two of the patient organisations addressed this policy option specifically, and expressed their support.[139] Member States who voiced their view on this policy option opposed it, stressing that the person who takes a risk should also be liable for damages arising from it and that Member States should not be put under an obligation to indemnify these damages.[140] The two national insurer's associations who responded to the public consultation were opposed.

1.1.35. Policy option No 2/6 — Combination of policy options No 2/4 and No 2/5

275.       This policy option is only relevant for the issue of obligatory insurance/indemnification, as policy option No 2/5 only addresses that aspect. As regards the annual safety reporting, the same assessment as in point 5.2.4 applies.

276.       Regarding insurance/indemnification it would mean that:

           For clinical trials with an IMP which is authorised, there is no insurance/indemnification which would apply to the general insurance coverages for the treatment of the patient outside a clinical trial (see point 5.2.4.1);

           For clinical trials with an IMP which is not authorised, the insurance/indemnification is ensured through the national indemnification mechanism.

1.17.1.11. 5.2.6.1.            Social/health impact

277.       The impact is the sum of the impact of policy options No 2/4 and 2/5, i.e. with regard to clinical trials with authorised medicinal products damages caused by the clinical trial are extremely rare, and would be covered by other types of insurance that cover the treatment of the patient (see point 5.2.4.1). For other clinical trials the national indemnification mechanism would provide coverage (see point 5.2.5.1).

1.17.1.12. 5.2.6.2.            Economic impact/costs

278.       The impact in terms of costs would be, as regards the annual safety report, identical to policy option No 2/4, as policy option No 2/5 does not concern the annual safety report.

279.       Regarding the obligatory insurance/indemnification this policy option reduces the administrative costs created by policy option No 2/5 by an additional 30 870 EUR (see Annex 4). This means that the savings compared to the baseline option created by policy option No 2/5 (1.66m EUR) are increased by the additional savings of 0.03m EUR to 1.69m EUR.

280.       The costs for covering damages, as well as implementation costs, would be identical to policy option No 2/5: the clinical trials addressed in policy option No 2/4, pose, if any, a minimal risk to subject safety and other insurance scheme are in place to address these potential damages (see point 5.2.4.1). The national indemnification mechanism would in practice only concern clinical trials which do not fall within policy option No 2/4.

1.1.36. Comparison of policy options for objective No 2 and synergies

281.       Policy option No 2/1 is not satisfactory, as it does not address the problem identified in point 2.2.2. Policy options No 2/2 to No 2/5 should be discussed instead, as they all offer an effective means to address the issue. In particular, all four other policy options offer significant savings (economic impact).

282.       Out of these policy options, No 2/2 has one major drawback: the ‘shifting back’ to Member States of the powers to regulate low-risk clinical trials means that these clinical trials are excluded from any harmonisation of clinical trials at EU level. However, as explained in point 2.2.1.3, it is crucial to facilitate pan-European clinical research with pharmaceuticals in order to address the requirements for sufficiently powered clinical trials.

283.       Policy option No 2/3 has been discussed widely in recent years. Today, there is strong consensus not to exempt ‘non-commercial sponsors’ as such from regulatory requirements. It is difficult to see why rules designed to protect the safety and rights of participants and the reliability and robustness of data should apply to some types of sponsor but not to others. Besides, it is difficult in practice to establish whether a sponsor is acting in a ‘non-commercial’ or a ‘commercial’ capacity. Commercial use of clinical trial data could be indirect or might not become apparent until after a clinical trial has ended.

284.       Both public consultations strongly supported this view.

285.       Moreover, there is an issue similar to that discussed for policy option No 2/2: if clinical trials by ‘non-commercial sponsors’ were excluded from the scope of the Clinical Trials Directive, they would not be subject to harmonised rules at EU level. Member States would again be responsible for regulating these trials via national laws. This would introduce differences in protection of trial subjects in the EU and would also make it more cumbersome to conduct such studies in the EU, which is not in the interest of ‘non-commercial sponsors’ performing clinical trials in different Member States.

286.       Policy option No 2/4 brings about less savings for sponsors than policy option No 2/3. However, in terms of public health and patient safety it is superior to policy option No 2/3. It leaves aside any differentiation between ‘non-commercial’ and ‘commercial’ sponsors (which is a key weakness of policy option No 2/3) and focuses on an objective criterion: the authorisation status of the IMP. Therefore, in comparison with policy option No 2/2 and No 2/3, policy option No 2/4 seems to be the best way to achieve the objective.

287.       Policy option No 2/5 is an issue apart. It addresses only the specific point of obligatory insurance/indemnity. It can offer synergies with policy options No 2/2 to 2/4. As shown above, policy option No 2/5 can be a useful tool to address the specific issue of obligatory insurance/indemnity. It greatly reduces the costs for indemnification by limiting contributions to the costs that actually occur: While, today costs for insurance/indemnification coverage are approximately 75m EUR per year, these costs would be limited to less than 1m EUR, thus addressing the concern of the very high insurance/indemnification coverage created by the Clinical Trials Directive. Policy option No 2/5 also greatly reduces the administrative costs and compliance costs for sponsors and yet provides protection for patients at least as strong and efficacious as the baseline option. These clear benefits of policy option No 2/5 come with the prize: Member States will have to set up a national indemnification mechanism which brings about complications and – in particular in the start-up phase – costs for resources. However, regarding these resources needs policy option No 2/5 allows for passing on the costs to the sponsors who benefit from the national indemnification mechanism. Even where these costs are passed on to the sponsor, policy optoion No 2/5 is still considerably cheaper than the baseline option. This policy option is therefore an not only an effective, but also an efficient means to address the challenges in terms of costs and complexities for sponsors to ensure coverage of subjects in terms of compensation for damages.

288.       Policy option No 2/6 combines the options No 2/4 and No 2/5. It brings about the savings of both policy options taken together while not compromising patient's rights and safety, and reliability of data generated in a clinical trial.

Overview — Impact of policy options to address objective No 2

|| Contribution to addressing the problem (+++=very important contribution; ++=important contribution; +=some contribution; o=no contribution) || Health/social impact compared with baseline (+++=very positive impact; ++=positive impact; +=some positive impact; o=no impact; -=negative impact) || || Economic impact/costs (in EUR) compared to baseline || Other comments

Other compliance costs || Administrative costs

Policy option No 2/1 (baseline) || (o) || (o) || Insurance/ indemnity || 0 || 0 ||

ASR || 0 || 0

Policy option No 2/2 (enlargement of the scope of non-interventional trials) || (++) || (--) || Insurance/ indemnity || - 4 m || - 0.12 m || Member States might again introduce regulation at national level (no level playing field).

ASR || - 13.1 m || - 0.1 m

Policy option No 2/3 (excluding "non-commercial  sponsors") || (++) || (--) || Insurance/ indemnity || - 14 m || - 0.5 m || See policy option No 2/2.

ASR || - 59.9 m || - 0.44 m

Policy option No 2/4 (removing requirements on the basis of the knowledge of the IMP) || (++) || (=) || Insurance/ indemnity || - 7.8 m || - 0.25 m ||

ASR || - 26.1 m || - 0.2 m

Policy option No 2/5 (insurance/indemnity) || (+++) || (++) || Insurance/ indemnity || - 74.2 m || - 1.66 m || Implementation needs per year: 700 000 EUR (=10 FTEs) Damages to be paid per year: 135 000 EUR Synergies possible with policy options No 2/2 to No 2/4

Policy option No 2/6 (combination of policy option No 2/4 and 2/5) || (+++) || (++) || Insurance/ indemnity || - 74.2 m || - 1.69m || As in policy option No 2/5

ASR || - 26.1 m || - 0.2 m

1.18. 5.3.    Objective No 3: Addressing the global dimension of clinical trials when ensuring compliance with GCP

1.1.37. Policy option No 3/1: Leaving the situation as it is (baseline option)

289.       This policy option would not address the pressing questions raised under point 2.2.2. The existing measures have flaws which make it difficult to achieve the envisaged aim. In particular, the existing voluntary self-commitment by most sponsors to perform clinical trials in accordance with GCP does not necessarily give all the guarantees needed and is not enforceable. The same could be true, depending on the country, of the existing regulatory supervision and inspections by non-EU countries in their jurisdictions.

290.       EU inspections are already performed in non-EU countries today, but only to a limited extent. Today, the EMA triggers, in the framework of the authorisation procedure of medicinal products, approximately 30 inspections outside the EU per year (see Annex 2). This rather limited activity is due to two factors:

· Resources: ‘Triggered inspections’ by the EMA are not performed by ‘EU staff’, but by the staff of NCAs of the Member States on behalf of the EU.[141] For each inspection, the Agency has to enquire which Member States have resources available. However, resources in Member States are increasingly limited;

· Ex-post assessment: Inspections in non-EU countries suffer from one major limitation: they are performed years after the clinical trial has ended and thus limited to ex-post verification of archived documentation. This is less effective than inspection of an ongoing clinical trial.

1.1.38. Policy option No 3/2: Facilitating GCP inspections by increasing transparency

1.18.1.1. 5.3.2.1.          Social/health impact

291.       This policy option would contribute to securing compliance with GCP with the aid of a stronger degree of transparency.

292.       Only if it is publicly known that a clinical trial is in progress can the control mechanisms of competent authorities (be they in non-EU countries or in the EU) be effective: In particular, inspections of GCP compliance can only be conducted if information is available as to whether a clinical trial is ongoing in a given country or not.

293.       Moreover, only transparency about the conduct of clinical trials can ensure effective scrutiny by media and civil society.

294.       It is for this reason that, during both public consultations, all stakeholders supported this policy option.

1.18.1.2. 5.3.2.2.          Impact in terms of costs for sponsors

295.       The impact on costs for sponsors will mainly be felt in the administrative costs for submitting information on clinical trials in non-EU countries to a public register.

296.       Details of these administrative costs are given in Annex 5. They total approximately 6.72 m EUR per year. These costs are administrative burdens, as they would not arise otherwise.

1.18.1.3. 5.3.2.3.          Implementation costs

297.       Implementation costs would be minimal. As set out above, a publicly-accessible register for clinical trials exists already at EU-level. Allowing upload of information on clinical trials performed exclusively in third country is a very simple IT exercise.

1.1.39. Policy option No 3/3: Inspections of non-EU countries' regulatory systems for clinical trial

1.18.1.4. 5.3.3.1.          Social/health impact

298.       This policy option would focus on the system of supervision of enforcement in non-EU countries. As a consequence, this policy option would not duplicate national/local inspection mechanisms in non-EU countries. In doing this, this policy option would contribute to securing compliance with GCP in clinical trials performed in non-EU countries.

1.18.1.5. 5.3.3.2.          Economic/resources impact

299.       The main costs would stem from resource needs at EU level. To date, there are no such resources allocated at EU-level.

300.       Annex 5 describes the resources required for 'system inspections' (see point 4.3.3), based on the experience in the food and veterinary sector, where a system similar to that put forward in this policy option exists (Food and Veterinary Office of the Commission, FVO). It shows that, to conduct approximately 8 system inspections per year of regulatory/supervisory systems, 3 inspector-FTEs plus 2 support-FTEs are needed. Additional implementation costs would be approximately 76 000 EUR.

301.       For the purpose of this impact assessment it can be left open whether the inspection activity set out in this policy option would be allocated with the Agency (who has experience in coordinating GCP inspections in the context of the EU-wide marketing authorisation procedure) or with the Commission, who could draw on experiences from the area of food and veterinary control. This is a political decision to be taken at a later stage, to which the impact assessment report shall serve as an aid for decision making (see Annexes 5 and 8).

1.1.40. Policy option No 3/4: GCP inspections of non-EU countries' clinical trial sites

1.18.1.6. 5.3.4.1.          Social/health impact

302.       This policy option would contribute to securing compliance with GCP in clinical trials performed in a non-EU country. However, it is not such a powerful tool as it might seem, mainly for the following reasons:

· Extent of clinical trial activity: As indicated in Annex 2, the majority of clinical trial results submitted in pivotal clinical trials in a marketing authorisation procedure stem from non-EU countries. Between 2005 and 2009, these pivotal clinical trials were spread over 44 034 sites in 89 countries.[142] It is impossible to inspect all these sites regularly and systematically.

· Inspections are usually conducted in the context of the marketing authorisation procedure, i.e. many years after the clinical trial has ended. Because of this lapse in time it can in some cases be difficult to assess with certainty whether the clinical trial was conducted in accordance with GCP.

1.18.1.7. 5.3.4.2.          Economic/resources impact

303.       The main costs would stem from resource needs for at EU level. As for 'system inspections' (see point 5.3.3), there is currently, at EU-level, no inspection capacity foreseen.

304.       In view of the sheer number of clinical trial sites in third countries, a systematic inspection of all relevant clinical trial sites would be unfeasible. Therefore, it is assumed here that 10% of all clinical trial sites contained in pivotal clinical trials would be chosen for inspection on the basis of risk-criteria. Even in this case, however, resource needs would be in the range of 1 300 FTEs at EU-level (see Annex 5).

1.1.41. Policy option No 3/5: Combination of policy options No 3/2 and 3/3

1.18.1.8. 5.3.5.1.          Social/health impact

305.       The combination of policy options No 3/2 and No 3/3 would have a further strengthen the favourable impact on GCP compliance in clinical trials performed in non-EU countries. This is because the transparency (policy option No 3/2) as to where clinical trials relevant for the EU are conducted allows targeting inspections on non-EU countries' regulatory systems.

1.18.1.9. 5.3.5.2.          Economic/resources impact

306.       The impact in terms of costs and resources is the cumulative impact of policy options No 3/2 and 3/3.

1.1.42. Comparison of policy options for objective No 3 and synergies, subsidiarity

307.       The foregoing shows that policy option No 3/1 is not satisfactory: it would not address the problem identified under point 2.2.3.

308.       Regarding policy options No 3/3 and No 3/4, both have relatively similar effects in terms of achieving the objective, even though the approach is different. Their impact diverges considerably as regards the impact on resources at EU level. Regarding policy option No 3/4, the budgetary cuts, in particular in personnel at EU level (both the Commission and the Agency), do not allow, at present, an increase in inspection activity in line with policy option No 3/4. The assessment of the impact of policy option No 3/3 shows that much can be achieved with far fewer resources than specified in policy option No 3/4.

309.       Policy options No 3/3 and 3/4 are not per se mutually exclusive: One could conduct inspections of regulatory systems and inspections of trial sites. However, there are, in practice, resources limitations which would not allow a cumulative application of both policy options: The resources available at EU level would not allow the effective conduct of both activities in parallel.

310.       Therefore, while these two policy options do not exclude each other in theory, in view of the limited resources they cannot apply, in practice, cumulatively.

311.       Policy option No 3/2 can make a useful contribution to effective control over clinical trials performed in non-EU countries. The burden for the sponsor, which is limited to administrative costs, is acceptable in view of the benefits created by this policy option.

312.       Policy option No 3/5 combines policy options No 3/2 and No 3/3: This combination of policy options is preferable as it strengthens further the favourable impact of policy option No 3/3. In terms of subsidiarity considerations, the preferable policy option No 3/5 is based on the exercise of a shared competence provided for in the TFEU to regulate the conduct of clinical trials. As set out above (point 2.3), this competence has been exercised by the Union legislator and consequently Member States are not allowed to add on these rules. This also holds for the rules as to whether data generated in clinical trials is acceptable in the context of authorisation of a clinical trial in the EU or authorisation of a medicinal product authorised in the EU. Policy option No 3/5 addresses this point: The data submitted in the EU has to stem from clinical trials which are conducted in accordance with GCP, and which are publicly registered.

             

Overview — Impact of policy options to address objective No 3

|| Contribution to addressing the problem (+++=very important contribution; ++=important contribution; +=some contribution; o=no contribution) || Health/social impact compared with baseline (+++=very positive impact; ++=positive impact; +=some positive impact) || Economic impact/costs compared with baseline || Other comments

Policy option No 3/1 (baseline) || (o) || || n/a ||

Policy option No 3/2 (increased transparency on GCP/ obligation of registration for all CT) || (++) || (++) || Additional administrative burdens for sponsors: 6.72 m EUR || Synergies possible with policy options No 3/3 and 3/4

Policy option No 3/3 (Inspections of regulatory systems in third countries) || (++) || (++) || Additional resource needs at EU level: 5 FTEs, 76 000 EUR ||

Policy option No 3/4 (GCP sites inspections in non-EU countries) || (++) || (++) || Additional staff needs at EU level: 1300 FTEs ||

Policy option No 3/5 (Combination of policy options No 3/2 and No 3/2) || (++) || (+++) || Cumulative impact of policy options No 3/2 and No 3/2 ||

6.           Conclusion — Final choices of policy options — Overview

1.19. 6.1.    Final choices of policy options

313.       Comparison of the impact of the policy options in Chapter 5 leads to the following conclusions:

314.       Concerning objective No 1, the comparison in point 5.1.6 shows that the baseline option No 1/1 is insufficient. Comparing policy options No 1/2 (single submission with separate assessment) to No 1/4 (single submission with central assessment by the Agency), policy option No 1/3 (single submission with joint MS assessment) has the best arguments on its side: it reduces the administrative costs by over 270m EUR. These costs are all administrative burdens. Moreover, this policy option leads to a fast approval procedure without recourse, as in policy option No 1/4, to complex approval infrastructure at EU level, which would increase costs and delays and would be too burdensome especially for academic research. Moreover, the ‘institutional continuum’ between trial approval and medicines approval would pose the risk that clinical trials could be approved on the basis of data desirable for marketing authorisation, rather than from the point of view of the benefits and risks to the subject. Policy option No 1/5 is very effective and has particular benefits if it is applied, as in policy option No 1/6, as an add-on to policy option No 1/3 in order to streamline approval procedures.

315.       Concerning objective No 2, the comparison in point 5.2.7 shows that excluding ‘non-commercial’ sponsors from the scope of EU regulation of clinical trials (policy option No 2/3) would be the wrong approach. Equally, reducing the scope of EU regulation (policy option No 2/2) would be counter-productive, as clinical trials excluded from the scope of EU law would be regulated at national level, which runs counter to the interests of public health and sponsors. Instead, policy option No 2/4 (removing requirements on the basis of the knowledge of the IMP) offers a viable solution to achieve objective No 2 without compromising public health, patient safety or harmonisation efforts at EU level. For the specific topic of obligatory insurance/indemnity, policy option No 2/5 is very effective as an add-on measure in order to achieve objective No 2. In view of this, a policy option No 2/6 combining policy options No 2/4 and 2/5 is best achieving the objective No 2.

316.       Concerning objective No 3, as indicated in point 5.3.6, some of the policy options can be added together and are not mutually exclusive. When assessing the options, costs must be considered which could affect not only sponsors and investigators, but also EU institutions (the Commission or the Agency) and thus, indirectly, the European taxpayer or sponsors as fee-payers. The conclusion is that, by joining policy options No 3/2 (increased transparency on GCP/ obligation of registration for all CT) and 3/3 (Inspections of non-EU countries' regulatory systems for clinical trials) together, the objective can be achieved reasonably well without recourse to policy option No 3/4 (GCP inspections in non-EU countries' clinical trial sites). In view of the mutually-strengthening effect, a policy option No 3/5, combining policy options No 3/2 and 3/3, is the best way to achieving objective No 3.

1.20. 6.2.    Final overview

Overview of the impact of the final policy choices

Chosen policy options No || Contribution to addressing the problem (+++=very important contribution; ++=important contribution) || Health/social impact (+++=very positive impact; ++=positive impact) || Economic impact/costs compared with baseline (in EUR) || Other comments (including implementation costs)

Other compliance costs || Admininistr. burdens

1/6 (single submission with joint MS assessment & Regulation) || (+++) || (+++) || - 440 m || - 271.7 m || Single submission point: One-off: between 1.62m (limited IT solution) and 6.3m (extensive IT solution) Running: between 0.34m (limited IT solution) plus 0.25 FTEs and 1.26m plus 19 FTEs (extensive IT solution) Additional support staff needed: 1.75 or 7 FTEs (depending on choice for limited or extensive support structure)

2/6 (removing requirements on the basis of the knowledge of the IMP & insurance/indemnity mechanism) || (++) || (++) || - 100.3 m || - 1.89 m || -

3/5 (increased transparency on GCP/ obligation of registration for all CT & system inspections) || (++) || (++) || - || + 6.72 m || Additional resource needs at EU level: 5 FTEs, 76 000 EUR

7.           Monitoring and evaluation

317.       Once adopted, implementation and compliance of Member States with the revised legislation is going to be monitored under the auspices of the Pharmaceutical Committee. In accordance with the Council Decision establishing the Pharmaceutical Committee,[143] this body is tasked to examine any question relating to the application of Directives on medicinal products which are put forward. To this end, the Pharmaceutical Committee is composed of senior experts of the competent authorities of all Member States.

318.       Evaluation of the impact of the revised legislation is going to be based, in accordance with the operational objective set out above (point 3) on the following criteria:

· Development of the number of clinical trials applied for in the EU, as well as the number of clinical trial participants; and

· Development of the number of multinational clinical trials applied for in the EU.

319.       Both impact indicators will be retrieved on a regular basis from the single EU Portal (see point 4.1.3), which is going to hold this information. A compilation of these data is going to be published on a yearly basis.

320.       Moreover, evaluation of the impact is going to assess the following criteria:

· Development of the delays between finalisation of the protocol and 'first patient in';

· The development of administrative costs presenting administrative burdens, and of other compliance costs of clinical trials conducted in the EU; and

· Trends in conducting clinical trial outside the EU for generating data referred to in the request for authorisation of a clinical trial or a medicinal product.

321.       For the purpose of this evaluation, the Commission continues to be in constant contact with stakeholders associations (industry and non-commercial sponsors) through attending workshops and other relevant events.

322.       The impact indicator "compliance with GCP of clinical trials conducted in non-EU countries" is going to be measured through periodic evaluation of the results from inspections referred to in policy option No 3/3.

323.       The information gathered from these sources is going to be compiled in a comprehensive interim evaluation which will be made availale to the public five years after the date of application of the revised legislation.

324.       This comprehensive interim evaluation shall serve as basis for a stakeholder meeting similar to the Commission/Agency clinical trial conference, organised by the Commission is envisaged. In addition, a public consultation is going to be held.

325.       On the basis of these additional findings a final evaluation report is going to be published seven years after the date of application of the revised legislation.

[1]               COM(2008) 666 final.

[2]               OJ L 121, 1.5.2001, p. 34.

[3]               COM(2010) 623; see Annex II (point 20) and Annex III (point 41).

[4]               COM(2011) 777, see Annex I (point 54) and Annex II (point 10).

[5]               Established by Council Decision 75/320/EEC of 20 May 1975 setting up a pharmaceutical committee (OJ L 147, 9.6.1975, p. 23).

[6]               http://ec.europa.eu/transparency/regexpert/detail.cfm?ref=1464.

[7]               EMEA/565466/2007: http://www.eortc.be/services/doc/EUCTD/EC-EMEA_report_CT_20071003.pdf.

[8]               http://ec.europa.eu/health/files/clinicaltrials/concept_paper_02-2011.pdf.

[9]               COM(2002) 704.

[10]             http://ec.europa.eu/health/human-use/clinical-trials/index_en.htm.

[11]             http://ec.europa.eu/enterprise/policies/smart-regulation/files/hlg_opinion_pharma_050309_en.pdf. The recommendations are based on the ‘EU project on baseline and reduction of administrative costs — Measurement data and analysis for the pharmaceuticals legislation priority area’, Final report (March 2009) (http://ec.europa.eu/enterprise/policies/smart-regulation/files/abst09_pharma_en.pdf).

[12]             http://www.esf.org/fileadmin/links/EMRC/FL_IDCT.pdf.

[13]                http://www.ebmt.org/2RelatedMeetings/EFGCP/Road%20Map%20Initiative%20for%20Clinical%20Research%20in%20Europe_Information.pdf.

[14]             http://patientpartner-europe.eu/en/home.

[15]             http://www.oecd.org/dataoecd/31/8/49344626.pdf.

[16]             http://ec.europa.eu/governance/impact/practice_en.htm.

[17]             http://ec.europa.eu/governance/impact/ia_carried_out/cia_2011_en.htm.

[18]             Article 2(a) of the Clinical Trials Directive.

[19]             For the purposes of this document, all references to the EU or EU Member States include the EEA or EEA Contracting States, unless indicated otherwise.

[20]             The term 'sponsor' or 'sponsored' means the responsability under which the clinical trial is conducted. It is not to be confused with 'funded' (a clinical trial might be funded by another body than the sponsor). See also Annex 1.

[21]             Source: EudraCT. When looking at clinical trial applications, the share of industry sponsors is 80 % (one clinical trial can imply up to 27 applications, depending on the number of Member States concerned).

[22]             Information on the types of clinical trials is contained in Annex 1.

[23]             Article 1(2). See also the Introduction to the ICH guidance on GCP.

[24]             Point 8 of the introduction to Annex I of Directive 2001/83/EC.

[25]             Sometimes referred to as 'academic sponsors'. Both terms are used interchangeably.

[26]             Medical Research Council, ‘Trials that have changed the world’               (http://www.mrc.ac.uk/Achievementsimpact/Clinicaltrials/index.htm).

[27]             See also Arduino Verdecchia, Silvia Francisci, Hermann Brenner, Gemma Gatta, Andrea Micheli, Lucia Mangone, Ian Kunkler, and the EUROCARE-4 Working Group, Recent cancer survival in Europe: a 2000–02 period analysis of EUROCARE-4 data, Lancet Oncol 2007: 8: 784–96; Henrike E. Karim-Kosa, Esther de Vriesa, Isabelle Soerjomataram Valery Lemmensa, Sabine Siesling, Jan Willem W. Coebergh, Recent trends of cancer in Europe: A combined approach of incidence, survival and mortality for 17 cancer sites since the 1990, European Journal of Cancer, 44 (2008)1345 –1389.

[28]             http://eudrapharm.eu/eudrapharm/welcome.do?selectedStaticLocale.languageCode=en

[29]             The scientific findings diverge: See, for example Robinson WR, Ritter J, Rogers AS, Tedjarati S, Lieberenz C, Clinical trial participation is associated with improved outcome in women with ovarian cancer, Int J Gynecol Cancer. 2009 Jan;19(1):124-8; Vist GE, Bryant D, Somerville L, Birminghem T, Oxman AD. Outcomes of patients who participate in randomized controlled trials compared to similar patients receiving similar interventions who do not participate. Cochrane Database of Systematic Reviews 2008, Issue 3. Art. No.: MR000009. DOI: 10.1002/14651858.MR000009.pub4.

[30]             See, for example 'Clinical Trials in Poland' PriceWaterhouseCoopers, November 2010 (http://www.pwc.com/gx/en/pharma-life-sciences/publications/clinical-trials-in-poland-2010.jhtml).

[31]             Report of the Commission/Agency clinical trial conference, p. 26.

[32]             http://ec.europa.eu/enterprise/admin-burdens-reduction/highlevelgroup_en.htm.

[33]             EudraCT.

[34]             ICREL was a longitudinal, retrospective, observational and comparative study to assess the impact of the Directive on the number, size and nature of clinical trials and on workload, resources required, costs and performance. Mean differences between 2003 (i.e. before the Clinical Trials Directive entered into force) and 2007 were assessed. Fuller details of the findings of the study can be obtained from: http://www.efgcp.be/icrel/.

[35]             ICREL, p. 130.

[36]             ICREL, p. 144.

[37]             ICREL, p. 132; See also point 2.2.2.

[38]             ICREL, p. 128.

[39]             Hearn J, Sullivan R, The impact of the ‘Clinical Trials’ directive on the cost and conduct of non-commercial cancer trials in the UK. Eur. J. Cancer 43:8-13, 2007.

[40]             Moulton B, Two years later: the impact of the EU CTD. Why research in Europe has declined since the implementation of the Clinical Trials Directive. Applied Clinical Trials. August 1, 2006.

[41]             van Vyve D, Meunier F, Facing the Challenges of the European Clinical Trials Directive: the European Organisation for Research and Treatment of Cancer perspective, European Oncology, 2008; 4; 1.

[42]             For an overview of the non-EU regions involved today in pivotal clinical trials submitted at EU level for marketing authorisation purposes, see Table 14 in Annex 2.

[43]             See, for example, 'Durg testing goes offshore' (CNNMoney, 8 August 2005) quoting various industry sources.

[44]             S. W. Glickman, et. al., Ethical and Scientific Implications of the Globalization of Clinical Research, N Engl J Med 360; 8 February 19, 2009, p. 816.

[45]             Response to the 2011 public consultation, cover letter, page 2.

[46]             Articles 6, 9 of the Clinical Trials Directive.

[47]             In the context of ECs, the Clinical Trials directive uses the term 'favourable opinion'.

[48]             Article 6(5) and Article 9(4) of the Clinical Trials Directive.

[49]             The role of the Agency is limited to maintaining the EU clinical trials databasee EudraCT, and coordinating GCP inspection activity in the centralised authorisation procedure for medicines.

[50]             See for example the Conference report, section 3.2.2: "The burden of paperwork should be reduced by rationalising the application forms and the content of dossiers and by reducing the number of times the same or nearly the same information has to be submitted to different NCAs and ethics committees."

[51]             See point 2.2.

[52]             See for example the response of the Association of Clinical Research Organizations (ACRO) to the 2009/10 public consultation (p. 2): "The different application of the regulatory framework by Member States does not, in the majority of cases, lead to divergent decisions on clinical trial applications. However, prior to reaching the final decision, the questions raised by the national comjpetent authorities on the identical scientific dossier are frequently very different in both number and nature and indicate a significantly divergent approach to dossier assessment."

[53]             See, for example, the response from The European Clinical Research Organisation Federation in the 2009/10 public consultation (p. 3): "Almost for every international study the list of deficiencies for the Investigational Medicinal Product Dossier varies considerably between the different national competent authorities involved."

[54]             This figure is based on responses to the 2009/10 public consultation and has been double-checked in the in the 2011 public consultation (see page 22 of the public consultation document).

[55]             Response of ECPC in the 2009/10 public consultation, p. 5.

[56]             Response of EFPIA in the 2009/10 public consultation, p. 4.

[57]             Response of Roche in the 2009/10 public consultation (p. 3): "Thus even simple adjustments needed based on local requests from one country results into amendments ot the application in other countries where the clinical trial application is already approved."

[58]             For example the European Federation of Pharmaceutical Industries and Associations (EFPIA) stressed, in its response to the 2009/10 public consultation (p. 3): 'Experience of multi-country studies is that it is very unusual not to receive idvergent assessments and that these do lead to different requested changes in the protocol. This may lead either to a trial not being run in the Member State or to having to make multiple amendments to the protocol thereby delaying access to treatment for patients and increase in administrative burden and costs.'

[59]             Article 3(2)(f) of Directive 2001/20/EC. While the terms 'insurance' and 'indemnity' are not defined in the Clinical Trials Directive, for the purpose of this impact assessment they are to be understood as follows: 'Indemnity' is a broad concept entailing all mechanisms that are intended to compensate damages suffered by the damaged party. 'Insurance' is, more specifically, a mechanism whereby a third person guarantees payment of a compensation which is to be paid by the damaging party to the damaged party.

[60]             ICREL, p. 132.

[61]             Response of the Institut national de la santé et de la recherche médicale to the 2011 public consultation, p. 11.

[62]             Article 17(2) of Directive 2001/20/EC.

[63]             See for example the conference report, section 3.10.2 ('Final discussions and perspectives for the future – Non commercial sponsors'): 'A new legal framework should protecti participants according to the risk associated to the category of study, not to the study's commercial or non-commercial objective.'

[64]             Neaton JD, Babiker A, Bohnhorst M, Darbyshire J, Denning E, Frishman A, Grarup J, Larson G, Lundgren J., Regulatory impediments jeopardizing the conduct of clinical trials in Europe funded by the National Institutes of Health, Clin Trials. 2010 Dec;7(6):705-18. Epub 2010 Aug 20.

[65]             See point 1.7.

[66]             Point 8 of the introduction to Annex I of Directive 2001/83/EC. The introduction to Annex I of Directive 2001/83/EC has, as the entire Annex, legally-binding force as secondary Union law.

[67]             See also the study commissioned by the European Parliament - Directorate-General for External Policies of the Union: “Clinical trials in developing countries: How to protect people against unethical practices”, April 2009 - http://somo.nl/publications-en/Publication_3035/at_download/fullfile.

[68]             Final report of the expert meeting ‘Clinical trials and protection of trial subjects in low-income and developing countries’, Wemos, January 2008.

[69]             “Ethics for Drug Testing in Low and Middle Income Countries – Considerations for European Market Authorisations”, SOMO, February 2008.

[70]             “Ethical concerns in clinical trials in India: an investigation” of the Centre for Studies in Ethics and Rights, Mumbai, India., February 2009; http://www.fairdrugs.org/uploads/files/Ethical_concerns_in_clinical_trials_in_India_An_investigation.pdf.

[71]             Article 3(3) of Directive 2001/83/EC.

[72]             See the first citation in Directive 2010/84/EU (OJ L348, 31.12.2010, p. 74), Directive 2011/62/EU (OJ L174, 1.7.2011, p. 74) and Regulation (EU) No 1235/2010 (OJ L348, 31.12.2010, p. 1).

[73]             Article 6(1) of the Treaty on European Union.

[74]             'Administrative costs' are defined as the costs incurred by enterprises, the voluntary sector, public authorities and citizens in meeting legal obligations to provide information on their action or production, either to public authorities or to private parties (cf. European Commission Impact Assessment Guidelines, Part III, page 46).

[75]             'Administrative burdens' are administrative costs which are generated solely because of a legal obligation, i.e. it excludes administrative costs which an actor would have had anyway, even in the absence of the legislation (cf. European Commission Impact Assessment Guidelines, Part III, page 45).

[76]             For the purpose of this impact assessment, the term 'other compliance costs' shall be defined as costs for compliance with regulation, other than administriative costs (cf. European Commission Impact Assessment Guidelines, Part I, point 2.3, page 10).

[77]             COM(2010) 2020, 3.3.2010.

[78]             See also the Commission Communication on the 'Europe 2020 Flagship Initiative Innovation Union' (SEC(2010) 1161, 6.10.2010).

[79]             COM(2007) 630, 23.10.2007.

[80]             COM(2005) 535, 25.10.2005.

[81]             Commission Work Program 2011, COM(2010) 623, 27.10.2010, Annex III, point 41.

[82]             COM(2007) 23, 24.1.2007.

[83]             COM(2010) 543, 8.10.2010.

[84]             http://www.hma.eu/uploads/media/VHP_public_CBB_22_Dec_08___hk_jan12.pdf.

[85]             Cf. third paragraph of Article 288 of the TFEU.

[86]             Article 1(1) of Directive 2001/20/EC. The terms ‘non-interventional study’ and ‘non-interventional trial’ are used here interchangeably.

[87]             See the report of the OECD Global Science Forum - Facilitating International Cooperation in Non-Commercial Clinical Trials (2011), p. 9.

[88]             This viewpoint was recently reiterated by non-commercial sponsors in McMahon, Conway, MacDonal, McInnes, The unintended consequences of Clinical Trials Regulation, PLOS Medicine, November 2009, Issue 11: "We would favour a combined tactic of lobbying to simplify and 'regulatory retreat' and perhaps we could then look forward to more 'specific modality' exceptions for non-commercial trials in furutre legislation".

[89]             Some Nordic MS have already such indemnification system in place.

[90]             https://www.clinicaltrialsregister.eu/.

[91]             In addition, in accordance with Regulation 1901/2006 on medicinal products for paediatric use information on paediatric clinical trials contained in a paediatric investigation plan has to be uploaded in this register – even if these trials are performed exclusively in third countries.

[92]             COM(2010) 573, 19.10.2010.

[93]             OJ C83, 30.3.2010, p. 389.

[94]             Commission Communication on a "Strategy for the effective implementation of the Charter of fundamental rights by the European Union", COM(2010) 573, 19.10.2010, p. 6.

[95]             The Commission's proposal for the MFF 2014-2020 is not yet adopted. But the Communication from the Commission to the European Parliament, the Council, the European Economic and Social Committee and the Committee of the Regions "A Budget for Europe 2020" defines some core elements, COM(2011)500 final.

[96]             Implementation costs are the costs incurred by public authorities involved in implementation. These include, for example, human and infrastructure costs, plus enforcement costs (cf. European Commission Impact Assessment Guidelines, Part III, page 38).

[97]             COM(97) 369 final, 3.9.1997 (published in OJ C 306, 8.10.1997, p. 9).

[98]             Respone of EURORDIS-rare diseases Europe to the 2011 public consultation, p. 1.

[99]             Response of the Association of European Self-medication Industry (AESGP) to the 2011 public consultation, p. 1.

[100]            Response of the HMA to the 2011 public consultation, p.1.

[101]            This solution would include user validation functionalities, an IT helpdesk, a business support helpdesk, and operational support.

[102]            The system would be built on existing IT functionalities.

[103]            Provided no Member State invokes the 'opt-out', see point 1.1.8.

[104]            Response of the wellcome trust to the 2011 public consultation (p. 3).

[105]            Response of the European Genetic Alliances' Network (EGAN) to the 2011 public consultation, p. 2.

[106]            See for example the responses from EURDIS (p. 2), ECPC (p. 1) and European Patient's Forum (EPF, p. 2) to the 2011 public consultation.

[107]            See below, policy option No 1/4.

[108]            Response of the Bundesverband der Pharmazeutischen Industire (BPI) to the 2011 public consultation, p. 3.

[109]            Response of the HMA to the 2011 public consultation, p. 4.

[110]            Response of France to the 2011 public consultation, p. 2.

[111]            Support to operating process, process development and management, templates, training, working group support, occasional crisis issues, public relation.

[112]            Support to operating process, working group support

[113]            This reasoning applies only to medicinal products falling within the scope of the centralised marketing authorisation procedure (i.e. authorisation of placing on the market by the European Commission).

[114]            Point A.6 of the Declaration of Helsinki (2008). See also Article 3 of the Additional Protocol to the Oviedo Convention (2005): ‘The interests and welfare of the human being participating in research shall prevail over the sole interest of society or science.’

[115]            See, for example, the response to the 2011 public consultation by the European Network of Paediatric Research at the EMA (p. 1): "Despite the attractions of a central assessment, analogous to the system available for licensing, the majority of enpr-EMA respondents agreed that at present this would be unworkable in view of the national differences in clinical and ehtical practice."

[116]            Response of the Copenhagen University Hospital to the 2011 public consultation, p. 1.

[117]            See for example the response of EUCOPE (with a large share of members being SMEs) to the 2011 public consultation: "EUCOPE favours a single submission with a 'coordinated assessment procedure' and not a 'central assessment'. [… ] Very few clinical trials are rolled out in more than five or six Member States. A closely coordinated virtual assessment procedure supported by a very good IT infrastructure and incolving the relevant country experts may provide a pragmatic and fast solution".

[118]            Cf. response of EFPIA to the 2011 public consultation, p. 8.

[119]            Response of the European Genetic Alliances' Network (EGAN) to the 2011 public consultation, p. 2.

[120]            See the response of the UK to the 2009/10 public consultation (p. 8): "The UK believes that a Regulation, despite providing a common legislative basis for clinical trial regulation across the EU, will not fully overcome the differences in interpretation that currently occur between Member States both by sponsors and national competent authorities."

[121]            Response of the ZU/KL Leuven to the 2009/10 public consultation, p. 3.

[122]            Osborne, Edward, O'Callaghan, Running an international paediatric non commercial clinical trial, Archives of Disease in Childhood, 2009, 94, p. 729-733 (submitted by the authors as response to the 2009/10 public consultation).

[123]            Response of UK Cancer Research to the 2009/10 public consultation, p. 8.

[124]            See resonse of EFPIA to the 2009/10 public consultation (p. 22): "However, a Regulation that accomodates every Member State's national interests and requirements would be disastrous. For a Regulation to improve the situation, it must be written with the principle of risk adaptation foremost in mind."

[125]            See response of the Arbeitskreis Medizinischer Ethik-Kommissionen in der Bundesrepublik Deutschland to the 2009/10 public consultation (p. 7): "We are also afraid that a Regulation will result in a lower level of patient safety than the current level achieved in Germany."

[126]            Article 114(3) TFEU.

[127]            Article 168(1) TFEU.

[128]            +++=very important contribution; ++=important contribution; +=some contribution; o=no contribution.

[129]            +++=very positive impact; ++=positive impact; +=some positive impact; o=no impact; -=negative impact.

[130]            Response from the Koordinierungszentren für Klinische Studien (KKS Netzwerk) to the 2011 public consultation (p. 5): "This would mean if a risk based approach cannot be adopted, we would urge that the definition is widened."

[131]            Response from Pfizer to the 2011 public consultation (p. 5): "It should be made explicit that non-interventional trials are not covered by the revised clinical trials legislation. We suggest that the European Commission develop a separate legal regime for non-interventional trials."

[132]            See the response of HMA to the 2011 public consultation (p. 10) who "agrees not to modify the definition of non interventional trials […] but to proportionate requirements for clinical trials, on a risk-based approach."

[133]            Response of ECPC to the 2011 public consultation, p. 3. See also the responses of Parkinson's UK, p. 2.

[134]            There were, however, also other views: See, for example, the resonse of the Arbeitskreis medizinischer Ethik-Kommissionen in der Bundesrepublik Deutschland to the 2011 public consultation (p. 4): "As authorised drugs are available for use anyhow (usually without any special requirements) it is hard to understand why the proper monitoring and documentation of the treatment and its outcome should be 'penalized' by red take, insurance, approval by drug authorities and the like."

[135]            Article 107b of Directive 2001/83/EC.

[136]            See, for example, the response of the Sociedad Espanola de Farmacologia Clinica to the 2011 public consultation (p. 8): "Waiver from the periodic trial safety report obligation […]. This is redundant with the periodic safety report on the medicine that is already being submitted to the national competent authority of the marketing authorization holder."

[137]            In 2006 during phase I clinical trial of monoclonal antibody TGN1412 the subjects to which it was administered developed a severe inflammatory reaction with shock-like symptoms and systemic organs failure.

[138]            Response of the Deutsche Forschungsgemeinschaft (DFG) to the 2011 public consultation, p. 3.

[139]            Responses to the 2011 public consultation submitted by EATG (p. 4) and EURORDIS (p. 3).

[140]            Cf. response of  the MHRA (p. 6) to the 2001 public consultation.

[141]            Article 15(2) of Directive 2001/20/EC.

[142]            Cf. presentation by Mr F. Sweeney, EMA, at the international workshop on a draft reflection paper on ethical and GCP aspects of clinical trials of medicinal products for human use conducted in third countries and submitted in marketing-authorisation applications to the EMA, 6-7 September 2010, London.

[143]            Council Decision 75/320/EEC of 20 May 1975 setting up a pharmaceutical committee, OJ L 147, 09.06.1975, p. 23.