<!--:it-->Draft EMA guidelines on transparency in clinical trials: Protecting commercially confidential information<!--:-->
Thanks to Rosa Maria Torraco for collaborating on the English version of this article
The Italian version of this article has been published on June 1, 2022 on AboutPharma.com, within our bi-monthly column on the new Clinical Trials Regulation EU 536/2014.
On April 7, 2022, the EMA published draft guidelines on the transparency of information uploaded to the Clinical Trial Information System (“CTIS”) for the specific purpose of giving operational guidance on the management of commercially confidential information (“CCI”) and personal data (Guidance document on how to approach the protection of personal data and commercially confidential information in documents uploaded and published in the Clinical Trial Information System—”Guidance”).
This document, albeit still in draft form, appears to be key for sponsors, as it may provide them with important guidance in preparing documents to be uploaded to the CTIS, especially considering the potentially high impact of transparency regulations in clinical trials on the protection of CCI.
THE PORTAL AND DATABASE UNDER THE CLINICAL TRIALS REGULATION
Under the Clinical Trials Regulation (“CTR”), the EMA established a European portal (the CTIS) as the single entry point for submission of clinical trial data and documents and a database containing the data and documents submitted through the CTIS. The database houses all relevant information (from application to results) for each clinical trial, thus ensuring transparency of this information at the European level.
Indeed, Article 81(4) of the CTR states that the database is accessible to the public unless confidentiality for all or part of the data and information contained therein is justified on certain grounds, the main ones being: (a) the protection of personal data under Regulation (EU) 2018/1725 on the protection of individuals with regard to personal data processing by EU institutions; and (b) the protection of CCI, in particular taking into account the marketing authorization status of the medicinal product, subject to possible derogation when disclosure is in the public interest.
That said, there are some documents uploaded to the database that are not subject to publication. These documents are listed in Annex I to the Guidance and include but are not limited to the quality portion of the IMP dossier, requests for information and sponsor responses related to quality of the drug, and financial agreements.
All other documents are subject to publication, so special attention should be paid to any information in them that may be considered CCI and may therefore be redacted at the request of the sponsor.
In this article we will focus on the guidance provided by the draft Guidance in relation to the protection of CCI, focusing on the relationship between transparency and protection of personal data of trial participants related to a future contribution.
HOW TO PROTECT CCI IN CLINICAL TRIALS
First, the Guidance clarifies that, depending on the type of document uploaded, it may or may not be necessary to provide both a “for publication” and a “not for publication” version. In cases where both versions are required, such as for GMP documentation with a qualified person’s signature, these documents must be provided at the same time. Only the “for publication” version will then be published, and it should contain neither personal data nor information that would be considered CCI. The “not for publication” version of the document is the original unabridged version containing all the information required by the member state concerned to assess the trial. Therefore, it can contain both personal data—if necessary—and CCI.
According to the Guidance, sponsors have two main tools at their disposal to protect the CCI included in documents intended for publication.
MECHANISM FOR DEFERMENT OF PUBLICATION
The first tool is the opportunity to request a deferment of publication of one or more documents. The deferral mechanism was introduced to provide sponsors (and member states) with the option of delaying publication of clinical trial information in order to protect CCI.
When submitting the initial application, the sponsor has the opportunity to request deferral and, if doing so, must provide the reason for the deferral. The extent of deferral and the resulting publication timeline depend on the trial category selected. During evaluation of an initial application, the Reference Member State (RMS) may ask the sponsor to make changes to the deferral settings through a Part I request for information.
Deferral rules apply to a subset of documents, including the protocol, investigator’s brochure, and investigational drug safety and efficacy documents, as well as summaries of trial results (only for certain studies) and certain key characteristics (only for certain studies). After the deferral period has passed, the version of the documents “for publication” uploaded to the CTIS secure domain will be published.
Regardless of whether the sponsor requests a deferral or not, it is required to submit a version of the documents “for publication” and a version “not for publication.” Specifically for mandatory clinical trial documents submitted to CTIS, either in an initial application or during the life cycle of the trial, a “for publication” version must be provided regardless of whether a deferral of publication is requested.
REDACTING CCI FROM PUBLISHED DOCUMENTS
The other essential tool available to sponsors is redaction, i.e., excising CCI from documents intended for publication. This requires correctly identifying pieces of information that can be considered CCI.
HOW TO IDENTIFY CCI
The draft Guidance is intended to provide operational guidance for the purpose of uniformity when it comes to what may or may not be considered CCI within clinical trial data and documents uploaded to the CTIS both during the application process and during the life cycle of the trial.
For the purposes of the Guidance, any information contained in a clinical trial application or provided during the trial life cycle that is not in the public domain or publicly available and the disclosure of which may harm the legitimate economic interests or competitive position of clinical trial sponsors is considered CCI.
Based on this definition, to facilitate identification of CCI, it is suggested that a three-step review be carried out in cooperation with experienced professionals with expertise in the area of clinical research:
- Step 1: This involves ruling out the possibility that the information targeted to be cut is already in the public domain. If the information is already in the public domain, it cannot constitute CCI and cannot be redacted. Sponsors are encouraged to create their own lists of the most common websites and other places where information related to their drugs is generally made available (including their own websites, the website of the EMA, clinical trial registries, the registries of other regulatory authorities including those outside Europe, and scientific literature and articles).
- Step 2: If the information is not in the public domain, it should be established whether the information is innovative and whether its disclosure would undermine the economic interests or competitive position of the owner of the information. Information that, while not in the public domain, is a logical and common-sense consequence of information that is contained in publicly available documents will not be considered innovative. If this additional requirement is also met, the information may be considered CCI and be stricken from the record. If, on the other hand, it is determined that the information is not groundbreaking, participations may proceed to Step 3 to seek an exception.
- Step 3: This step applies only in exceptional circumstances when, despite the information being non-innovative, the sponsor believes that its disclosure could harm the economic interests or competitive position of the information owner.
In general, the following are not considered CCI: (i) data included in a clinical study report (CSR) once a marketing authorization has been granted or a marketing authorization application has been withdrawn (this is consistent with the indications of EMA Policy 0070 and related application experience); (ii) the main features of a clinical trial; (iii) the conclusion of Part I of the assessment report for the authorization of a clinical trial; (iv) the decision on the authorization of a clinical trial, or the substantial modification of a clinical trial and its results, including the reasons for temporary halt and early termination. In these types of documents, therefore, CCI can be removed from the “for publication” version of the document.
Extensive redaction of CCI is not expected in a “for publication” version of a document that—due to deferral—will be made public only several months/years after the end of the trial in the EU or at the time of publication of the trial results. In redacting documents for which deferral is requested/granted, sponsors should therefore consider which elements will be considered CCI at the time of publication. It is expected that most of the elements that are considered CCI at the time of submission, based on the progress of clinical development, will no longer be considered CCI at the end of the deferral period and therefore will not need to be redacted.
BALANCING THE VARIOUS TOOLS
The tools mentioned above (deferral of document publication and redacting CCI from published documents) need to be properly balanced by sponsors to ensure the greatest possible protection for any CCI included in documents related to a clinical trial.
The deferral mechanism was introduced specifically to reduce the need to extensively redact CCI from submitted documents by delaying their publication to a time little information will still be considered CCI. In contrast, sponsors should not request deferment in cases where documents can be published with minimal CCI redaction. If a deferral is granted, the sponsor should consider in advance what limited information might still be CCI and thus redacted at the end of the deferral period.
In addition, identifying CCI as early as the documentation drafting stage reduces the need to redact information, promoting transparency. Therefore, it is recommended that early in the development plan sponsors use the three-step review described above to identify CCI and that they then keep track of it as the product evolves and proactively minimize the distribution of this information in the clinical trial documentation to be filed. The use of CCI in clinical documentation should be limited to cases where such information is essential, relevant, and/or specifically required. Following this approach will reduce the effort required to prepare separate versions of documents for publication and minimize the need for redaction.
So, it is crucial to make a careful assessment of these tools to avoid falling into the trap of attempting to over-protect CCI, as when not properly limited and justified, doing so raises the risk of rejection—with counter-productive effects (think of the strict control mechanism for CCI provided by EMA Policy 0070).
Indeed, although the Guidance applies to documents published in the EU Clinical Trials Database, it is certainly relevant to the application practice that has developed in the meantime and the provisions of EMA Policy 0070 that manages CCI in the context of requests for access to documents included in dossiers for the marketing authorizations of drugs. Both instruments are designed to achieve maximum transparency for information about medicines used within the European Union.
 The other reasons are c) the protection of confidential communications between member states in connection with the preparation of the assessment report; d) ensuring effective supervision of the conduct of a clinical trial by member states.
 The annual report on the publication of CSRs for centrally approved products as part of the EMA’s clinical data publication activities (Policy 0070) confirmed that out of 1,308,244 published pages, only 134 contained redacted material, which is equivalent to 0.01% of the total published pages.