The preparation and submission to ECs and CAs, as established by art. 9 of Directive 2001/20/EC  for multi-national studies is a challenging duty for the sponsor. In fact, each country has its own rules and procedures to comply with, and unfortunately, even each EC can have their own, while an ethical standard among trial centres should be guaranteed at the same time. In this work, we described our experience of a multi-centre multi-national CTA in EU and non-EU countries with different cultures, languages, geographical and political frameworks, as well as different pharmaceutical systems and prescriptive habits . In fact, notwithstanding the implementation of EU Directives into national laws and the acknowledgment of the EU system and guidance, differences in regulatory frameworks and procedures when applying a multi-national trial still remain.
On the other hand, the framework of non-EU countries for paediatric clinical trials is not yet implemented, thus leading to legal obstacles during the preparedness of a trial. Noticeably, DEEP-2 has represented the first paediatric clinical trial carried out in Albania, where Law No. 9323/2004 on Drugs and Pharmaceutical Service  which defines the rules of pharmaceutical service and medicines production, trading, prescribing, quality control, and inspection, was not specifically addressed to clinical trials.
In fact, the preparation of the CTA for coordinating sites, as the first to be addressed, was longer in non-EU countries, while for EU countries the shortest time resulted in Italy (0,4 months), as expected because the sponsor was Italian with an already good knowledge of national rules, and the longest time was in Greece (21,7 months), given that it was required to translate the protocol in the national language.
Overall, the parallel submission was efficacious in the EU, with the exception of Italy where the CA authorisation was granted after the EC approval, causing consistent delays for trial initiation. In Italy, at the time of DEEP-2 submission and until very recently, while the single opinion was recognised by the national law D.lgs. 211/2003  implementing Directive 2001/20/EC , de facto the ‘satellite ECs’, in charge of the acceptance or refusal of the single opinion granted by the ‘coordinator EC’, were used to review the CTA documents in line with their own internal procedures. With regards to the CA, the legal Officer of the trial site was addressed after the ethics approval. In fact, while the above mentioned Directive  as well as the Italian law itself dealt with the ‘tacit authorisation’ - if the Competent Authority has not informed the sponsor of any objections within 60 days (with the mentioned exceptions) - in practice in Italy a trial may not start if the legal Officer has not provided the written authorisation! In fact, for many Italian satellite ECs the time necessary to complete the authorisation process was longer.
It should be underlined that, even if in the non-EU countries GCP provisions are not officially implemented in a national law, the type and number of documents required for the EC approval are similar to what required in Europe. This represents the most interesting aspect of the DEEP experience documenting that it is possible to perform a GCP-compliant paediatric trial also in a heterogeneous framework, following the EU regulatory guidance as reference.
Moreover, provided that all the legal specific requirements were met, the timing to obtain the final trial authorisation was significantly reduced in non-EU countries in comparison with the EU.
Regarding the EU, DEEP experience has also documented the need for simplifying and further harmonising the authorisation process.
In the EU, the provision for CA authorisation and EC approval timing is indicated in Directive 2001/20/EC  as 60 days for the single opinion release and 30 days for its acceptance. Overall, our analysis revealed that in the EU countries the timeframes for the authorisation were not compliant with this requirement and longer than in non-EU countries.
The issues that affected the whole duration of the CTA in such a multi-national trial, resulted in the – long – lists of documents for submission, the requirement applicable in Italy and the UK to use separate portals from EudraCT to carry out the CTA and prepare the CTA form, and the queries raised from CAs and ECs.
Differences exist regarding the documents for submission, notwithstanding the EU rules  and guidance [5;6].
Important issues for children’s protection were found in Italy: relevance of the trial and trial design, benefit/risk evaluation, foreseen risks, justification for including minors (if not given in the protocol) were required by law . A recent Italian law  has established that insurance for clinical trials in children must foresee at least 10 years from completion of the clinical trial as the minimum period of tail coverage for the risk; this is the minimum time required to ascertain their regular psychophysical development.
We demonstrated that queries impacted on the timing for getting the ECs opinion and the CAs authorisation in the EU. For example, the longest time for obtaining the full authorisation in Cyprus (32,9 months) was due to requests related with the protocol from the EC. In contrast, no additional queries were posed in non-EU countries where the approval process was quick and easy.
After the CTA of DEEP-2, some changes in the national regulatory frameworks occurred. For example, in Italy Laws n. 189/2012 (“Balduzzi Decree”)  and 158/2012  established the reduction of national ECs, and the Italian Medicines Agency (AIFA) became the national CA. Outside the EU, in Albania the first law specifically dealing with clinical trials and partially aligned with the EU Directive  was released in 2014 .
Well-known differences exist to collect the informed consent at national level . Even if all countries acknowledged the need for collecting the informed consent from parents or legal representatives, in Albania, Cyprus, Greece, Italy and Tunisia both parents must consent for the participation of a child in the trial (contextually or in different moments), while in the UK and Egypt the signature from only one parent is accepted. This issue would be impracticable to harmonise, even in the light of a common EU regulation.
Part of queries raised by ECs dealt with the informed consent/assent process. Our data show that the assent form was mentioned in the list of documents to be provided for the CTA in Italy, UK and Egypt, but in no country the assent form and information material for paediatric subjects is mandatory by law. Of note, Egypt was the only non-EU country to list the assent among the documents to be submitted.
On this specific topic, notwithstanding DEEP provided common ad hoc material for the informed assent (three different age-tailored informative booklets and two assent forms prepared with the contribution of the patient representatives, shared among partners and made available in six languages  based on the EC Recommendations , the position of the involved countries demonstrated to be different: in the UK, the age-tailored booklets prepared by the sponsor with the support of a communication team (one for pre-school children with coloured figures, one for school-age children with figures and friendly childish explanations, and one for adolescents with more complete information and images adjusted for this age range) were not accepted by the EC as considered too complicated for school children and not suitable for adolescents.
For other issues, very different points of view were expressed and therefore it was necessary to adopt ad hoc solutions to respect national social and religious habits. This was the case of contraception: two ECs in Italy required to include more details on this issue both in the documents for girls and for boys; the Greek EC required to cover contraception failures in the insurance coverage; in the UK, it was required to explain in the information sheet for adolescents what would happen if a young person was found to be pregnant; in Egypt, the ECs asked to avoid all details on contraception in the consent and assent documentation, since not considered tailored for girls.
All these issues represented a great challenge and resulted in a global delay for starting the authorisation process. These issues were solved through a strong interaction among investigators, DEEP coordination staff and ECs according to the DEEP project procedures.
In the UK, the two booklets for older patients (6-10 years and 11-17 years) were replaced by a single information sheets, In Egypt, the term ‘contraception was avoided in the information material for patients but considered in the oral consent procedure to parents.
So, what to do for improving this situation?
Ad hoc actions to support the acceptance of the Ethical Recommendations  provisions should be done, including educational and patient empowerment initiatives.
Improvement is expected from the already issued but not yet applicable CT Regulation  whose major aim is to overcome the drawbacks surrounding clinical research and the harmonisation among countries. However, this Regulation is not designed to affect the national ethics approval. Therefore, we will still need to know and address local requirements.
Some help could derive by the involvement of large research consortia able to deal with the differences and to set up a strong collaboration not limited to the scientific topics but including also cultural and social issues and the direct experience and contribution of patients.
Accordingly, networks on paediatric research acting at international or global level will be crucial. The first initiative to establish a global rare disease CTs network is under discussion at FDA level as a “Rare Disease Cures Accelerator” . By providing a more centralized infrastructure and common platform(s), this tool would support the conduct of CTs in rare disease populations.
All things considered and in spite of the wide regulatory, cultural and linguistic heterogeneity, this has represented an useful case study to highlight the existing differences and difficulties and also demonstrates that a GCP paediatric study is possible even in a heterogeneous context.