Investigator initiated clinical trials (IICTs): a systematic search in registries comparing Czech Republic and Portugal in terms of funding policies and scientific outcomes

Background: Clinical trials provide one of the highest level of evidence to support medical practice. Investigator initiated clinical trials (IICTs) answer relevant questions in clinical practice that may not be addressed by industry. This study aims to compare, for the first time, two European Countries in terms of registered IICTs, respective funders, scientific outcome and its impact on the national performance in IICTs. Methods: A retrospective systematic search of registered IICTs, over the past 13 years, using four clinical trials registries was carried out in two European countries with similar population and Gross Domestic Product (GDP) but with different governmental models to fund clinical research. Each completed IICT was screened for sponsors, funders, type of intervention and associated publications. Results: IICTs involving Czech Republic and Portugal were n=439 (42 % with hospitals as sponsors) and n=328 (47 % with universities as sponsors), respectively. Funding agencies (national and international) supported only 25 % and 18 % of the IICTs in Czech Republic and Portugal, respectively. The Czech Ministry of Health, through the Health Research Council, supported 61 IICTs and the Portuguese Ministry of Science, Technology and High Education, through Fundação para a Ciência e Tecnologia , supported 27 IICTs. Among these, trials with investigational medicinal products represent 52 % (n= 32/61) and 4 % (n=1/27) in Czech Republic and Portugal, respectively. The Czech Republic presented a higher percentage of IICTs´ publications in high impact factor journals (IF>21) with national investigators as authors, when compared to Portugal (86 % vs 15 %). Conclusion: We found a better performance of Czech Republic in terms of number and scientific outcome of IICTs when compared to Portugal. These findings might be related to the existence of specific and periodic funding for clinical research in Czech Republic, although further data still need to be collected to confirm this relationship.


Background
In the last decades, clinical research played a crucial role on increasing the medical knowledge for the prevention, diagnostic, treatment and cure of diseases. Clinical trials provide one of the highest level of evidence to support medical practice. Investigatorinitiated clinical trials (IICTs) are pivotal to generate relevant and unbiased data decisive for the implementation of new therapies and better-informed decisions by health regulators [1]. However, the importance and relevance of IICTs rely heavily on its independence and impartiality, which in turn is directly linked to the funding source. Some IICTs are funded by industry, which can introduce a bias in the outcome of these studies.
In fact, a systematic review of 1140 studies demonstrated that industry-sponsored clinical trials were significantly more likely to reach conclusions favourable to the sponsors than those who were not industry-sponsored trials [2]. Therefore, public national funding agencies have a critical role on supporting fully independent clinical research with the sole purpose of benefiting the large patientsṕ opulation through the improvement of medical knowledge. Public funding of IICTs can also play an important role on addressing important clinical questions that remain unresolved because they are not addressed in industry-funded trials. Crowe and colleagues showed that there is an important mismatch on clinical research priorities identified by the patients and the clinicians, which in part is affected by the commercial aspect of clinical research [3].
The unbalanced ratio between private and public funding of IICTs (and its impact on national health systems through the definition of public policies for the treatment of diseases) is particularly relevant in countries with negligible expenditure in this area, which is usually associated to limited funds for the promotion of clinical research. It is important to generate significant scientific data in order to be able to provide better and informed guidance for the funding of IICTs by the national public funding agencies.  Furthermore we collected and compared the national public policies implemented by the Portuguese and Czech governments to promote clinical research during this timeframe.

Search methodology for clinical trials identification in each database
Step 1. The search in the EU-CTR allowed to include as general term "non-commercial" therefore eliminating the commercial studies. The timeframe selected was from 01/01/2004 until 31/12/2017 as well as the countries, Czech Republic or Portugal. No other requirement in the advanced search field was selected.
Step 2. As the discrimination between commercial and non-commercial clinical trial is not possible at the ClinicalTrials.gov platform, an advanced search was performed. The IICTs were extracted by selecting the predefined options: "Interventional studies", the abovementioned time period, each one of the countries, restricting to: NIH (U.S. National Institutes of Health), Other U.S. Federal agency, all others (individuals, universities, organizations) and Industry were chosen.
Step 3. Registrations in the BiomedCentral CTR are associated with ISRCTN. In this CTR it was not possible to refine the search with the exception of selecting Czech Republic or Portugal as the recruiting country. Trials initiated before or after the referred timeframe were discarded.
Step 4. In the ANZCTR, we performed the search in the selected timeframe, using Czech Republic or Portugal as the recruiting country. Registrations in this CTR have the initial code ACTRN.
In steps 1-4 some studies were discarded, using as exclusion criteria: industry sponsored trials; starting date before 01/01/2004; sponsor or recruitment site not involving Czech Republic or Portugal; non-intervention trials. The search results were reviewed individually by two independent experts in each national team to confirm the compliance with the inclusion and exclusion criteria.

Data extraction, duplicates and complementary information
Step 5. Information about trial identification (Trial ID) number (main and secondary), recruitment status, sponsor name and country, trial title, trial phase (when applicable), type of intervention, therapeutic area, design characteristics, type of funding and publications of the completed studies, were extracted manually and independently from the registered records (from 1/03/2018 to 31/08/2018), gathered and organized in Excel sheets. Forty percent of the records were double-checked by two individuals from each team and discussed until a consensus was reached.
Step 6. Duplicates were identified in each database, when secondary IDs were provided or when the sponsor and the study title were the same. The same study registered in different CTRs was considered duplicated and removed at this point of the search. When complementary information was provided in different registrations about the same study it was added to the respective entry in the working Excel sheet. None of the studies identified had recruitment sites in both countries, so no clinical trial is considered in both databases.
Sponsors were coded as: Disease-Specific Organization (Disease associations or research institutes dedicated to a specific therapeutic area), Foundation, Hospital, Research Institutes (non-specific therapeutic area), University, and others (Private health clinic, Funder Agency, acting as sponsor). In the four CTRs, the funding source was identified in different fields. Both in ISRCTN and ANZCTR there is a specific field entitled "Funder". In Clinicaltrials.gov CTR the funders was identified from the field "Sponsor and Collaborators". In EU-CTR the funder was identified in the field "sources of monetary support". All the funders identified in these CTRs were classified as: i) Public organizations (non-profit organization such as public institutions, funding agencies, disease specific organizations); ii) Private organization (for-profit organization -Industry); iii) Both, when the funding is provided by the industry and one or more public organizations; iv) Not Indicated, when the information was not provided in any registry or in the publication.
When only the sponsor was mentioned in the ClinicalTrials.gov registry, we considered it as the funder. The support of funding agencies was perceived through secondary IDs where the code of the grant agreement is in some cases added. Exclusive funding for PhD or Post-Doctoral grants was not considered.

Search of publications of completed IICTs
Step 7. Information regarding the publications of the registered and completed IICTs were manually and independently screened. The Trial ID was used to search abstracts of journals indexed in Medline (using PubMed) as well as in the four CTRs. By the principal investigator´s name was possible to identify publications, when no paper was found with the previous strategies. In some publications, the Trial ID was not included in the abstract, which rendered the search more difficult. All completed trials and published until 31/12/2017 were considered for further analysis. Subsequently, screening of each publication was performed to complete the information about the funding sources of IICTs to identify any other relevant information that was not complete in CTRs. The journal's impact factor was obtained from Web of Science, Research Gate or Bioxbio.com, considering this order when different impact factors for the year of publication were found for the same journal. Around 40% of the data were mutually exchanged between Czech and Portuguese team and double checked by two independent reviewers.

Identification of non-commercial studies
The number of trials identified in each database is shown in Figure 1. After discarding industry-sponsored trials, non-interventional studies, or those with no recruiting sites in Czech Republic or Portugal, 485 and 378 non-commercial trials (i.e. IICTs) were respectively considered eligible from all the screened databases. The same trial registered in different CTRs was separated and considered as duplicate. After discarding duplicates, a total number number of 439 IICTs were identified involving Czech Republic whereas for Portugal 328 trials were found. Forty-one percent and 49% of the studies were already completed and from those 41% and 58% were published, in Czech Republic and Portugal, respectively.

Characteristics of IICTs in Czech Republic and in Portugal
In both countries, most of IICTs has a national sponsor and were performed in the respective country ( Figure 2A

Clinical trials funded by national and international funding agencies
One of the great public funders of the IICTs in both countries were funding agencies  (Table 1), 32 (52%) are testing medicinal products (data not shown).
On the other hand, among the 60 IICTs in Portugal receiving financial support from funding grant agencies, only half of these (n=30/60) were funded by national funding grant agencies or funding programs ( Figure 3B) while the other half was funded by international funding agencies. Among the 27 trials funded by Fundação para a Ciência e Tecnologia (Table 1) only one (4%) is testing medicinal products (data not shown). The level of participation of these two countries in international grants was similar (n=38 and n=30, respectively), most of these involving trials with IMPs in both cases ( Figure 3A, The organizations receiving national funds, thus acting as sponsor, in higher number of trials in Czech Republic were hospitals (32%: n=35/109) and universities (22%: n=24/109). Whereas in Portugal were the universities (40 %: n=24/60) that sponsored more trials funded by national funding agencies followed by hospitals (7 %: n=4/60).
Portuguese hospitals sponsored more trials funded by international funding grants then with national funding grant agencies (n=8 vs n=4).  Fig. 1 D,E). When compared to Czech Republic, a lower number of funded IICTs by public funding agencies were identified in Portugal, however, the national investigators have been involved in trials with more diverse funding opportunities, especially international ones.

National policies for clinical research
Both countries present national public policies implemented (or not) for clinical research, in particularly IICTs, which are summarized in Table 2 Nevertheless, to our best knowledge in Czech Republic, the higher education courses specific to train clinical trials professionals are being prepared. In Portugal, the national government supported and invested in specific training for medical doctors 14 and research team professionals from hospitals [14].   Figure 4B).
Furthermore, international funding agencies funded 71 % of trials in Czech Republic and only 31% in Portugal that were published in journals with IF>21 (Inset B1, Figure 4B).
None of these published trials in high impact journals was funded by the Portuguese national funding agency while in Czech Republic, one of these trials was funded by the national funding agency. Around half of the trials published in journals with IF>21 are funded by private funders (43% for Czech Republic and 54% for Portugal). Additionally, most of these published trials were multinational and focused in IMPs, in both countries. instruments (e.g Era-Net and Co-funding).

Discussion
In this study we show that the main funders of IICTs in both countries are public or nonprofit organizations ( Figure 2D). However, we cannot provide accurate data on the origin of funding because, as we previously published, CTRs do not provide a clear separation between sponsor and funders 5 . Only a small percentage of the public funding to IICTs in both countries is provided by national funding agencies (25% in Czech Republic and 18% in Portugal - Figure 3). In Portugal, national funding agency is mainly used to provide financial support to behavior studies ( Figure 3B) that are less challenging in terms of costs, regulations, guidelines compliance and impact when compared to clinical trials with IMPs. Only 1 of the 27 studies supported by FCT is testing medicinal products (4%). In contrast, international funding agencies mainly fund trials with IMPs in Portugal ( Figure   3B). This difference was not found in Czech Republic where trials with IMPs were funded by both national and international funding grant agencies (Figure 3 A). The health funding agency in Czech Republic (AZV) funded 32 studies testing medicinal products out of 61 (52%). We may attribute these findings to the fact that AZV launch calls more frequently and adequate to support trials with IMPs. During the time period studied in the present work FCT launched only one call, in 2007, specifically focused in clinical research. The Portuguese Ministry of Health had a financial contribution to this call and a total amount of 9 M € were used to fund 63 projects [20]. However, none of these funded studies were found in our search, probably because most of them were not interventional studies or interventional without IMPs which registration is desirable but not mandatory.
Nevertheless, screening the information provided by FCT we found at least one interventional study with IMP that was not properly classified, consequently not registered in a database by the investigators [21].
In Czech Republic, University hospitals are pressured to apply to grant support and carry The existence of national funding grant agencies is of major relevance to implement pilot IICTs that later could be escalated to the European level. Submission of a project proposal that include a multinational trial to a international funding grant agency that had already a national pilot might have a higher probability to get financial suport. Moreover, the national funding schemes would also contribute to empower investigators to internationalize their ideas and coordinate multinational collaboration consortiums to capture international funding.

The national governments and European Commission investment in clinical research
infrastructures have been critical for the development of this area [23,24]. Both, Czech [15] and the Clinical Investigator Certificate, CLIC [14] is an asset to trigger a high quality clinical investigation. However, strategies to allocate research time to clinical investigators still need to be implemented [8].
The publication that resulted from the IICTs with authorship of the national investigator as In this study, we observed a low number of multinational IICTs registered in both countries when compared to other countries [6]. This difference might be due to the specific rules of from the interests of any other stakeholders [28]. Therefore, it is critical for the states to promote IICTs through a comprehensive and stimulating set of public policies and funding.
For example, Italian government decided to invest on IICTs funding, through the Italian Medicines Agency (AIFA) that resulted in an increase of the number of independent clinical research and scientific outputs [29,30]. This initiative started in 2005 and an amount of 40M€/year have been provided by the industry to AIFA, to fund competitive IICTs projects developed by Italian physicians working in the public/non-profit sector [30].
As a limitation of this study, the number of IICTs with IMPs in both countries might be underestimated due to lack of registration and/or misclassification, although being

Consent for publication
Not applicable. The data used in this work was obtained from public clinical trials databases.

Acknowledgments
The

Funding
In Czech Republic, this work was supported by the European Regional Development

Availability of data and materials
All data generated or analysed during this study are included in this published article [and its supplementary information files]. After discarding commercial trials all remaining studies were gathered in one