Since 31 January 2023, every new clinical trial application for a medicinal product in the European Union has been submitted through CTIS. Regulation 536/2014 sets out what the dossier must contain, and the clinical trial protocol you attach is read by several assessors, in several Member States, against a tight timetable. An inconsistency between the protocol, its synopsis and the investigator's brochure will not sink an application, but it does trigger a request for information, and that is counted in weeks. Here is what the text requires, what assessors look at, and the checks worth running before you submit.

What Regulation 536/2014 expects from a clinical trial protocol filed on CTIS

Article 25 refers to Annex I for the content of the application dossier. The protocol sits in section D of that annex, points 14 to 24. Point 14 states the expectation: the protocol describes the objective, design, methodology, statistical considerations, purpose and organisation of the trial. Point 15 requires it to be identified by the trial title, the EU trial number, the sponsor's protocol code number, the version number and date, the short title, and the sponsor's name and contact details. Point 16 asks for an accessible, searchable format rather than a scanned image. Point 17 sets out the expected content: a summary of the relevant work, an assessment of benefits and risks, justification of the dosage, description of the study population, statistical methods and sample size calculation, inclusion and exclusion criteria, recruitment and consent arrangements, safety monitoring and data protection. Point 24 adds a document still forgotten: the protocol shall be accompanied by a synopsis.

The protocol is more than a dossier component. Article 47 requires the sponsor and the investigator to conduct the trial in compliance with the protocol and with good clinical practice. In the European Union, the principles and Annex 1 of ICH E6(R3) came into effect on 23 July 2025. Our page on clinical trials and ICH-GCP sets out that scope.

Part I and Part II: two reading grids for the same text

The assessment is split. Part I, coordinated by the reporting Member State under Article 6, covers scientific and technical matters: expected health benefits, risks and inconveniences for participants, compliance with manufacturing and import requirements for investigational and auxiliary medicinal products, labelling, and the completeness of the investigator's brochure. Part II, assessed by each Member State concerned under Article 7, covers national matters: participant information and consent, compensation arrangements, recruitment arrangements, suitability of investigators and trial sites, data protection, and the collection and storage of biological samples.

The consequence is practical. The same sentence in the protocol is read by two assessors who do not have the same question in mind. The recruitment paragraph will be compared with each country's information documents. The benefit and risk paragraph will be compared with the investigator's brochure. Read the protocol twice, once against each grid.

The three documents that must say the same thing

The protocol, its synopsis and the investigator's brochure (section E of Annex I) describe the same trial at three levels of detail. Discrepancies almost always come from late amendments. Check them one by one:

  • the population: inclusion and exclusion criteria worded identically in the protocol and the synopsis, with no rewriting;
  • the dosage, route of administration, treatment duration and follow-up duration;
  • the primary endpoint, its definition and the time point at which it is measured;
  • the planned sample size and the statistical assumption behind it, including after a revised power calculation;
  • the reference safety information in the investigator's brochure, which determines whether a serious adverse reaction counts as expected;
  • version numbers and dates, on every cover page and in the application form.

A contraindication added after scientific advice, a visit window widened late in drafting, a sample size revised upwards: these three amendments are routine, and each one propagates into at least four documents.

What triggers a request for information

  • version numbers that disagree between the application form, the cover page and the body of the protocol;
  • a protocol supplied as a scanned image, contrary to point 16 of section D;
  • a missing synopsis, or one older than the protocol it summarises;
  • participant-facing documents drafted in a language the Member State concerned does not accept, when Article 26 leaves each State to set its own language requirements;
  • a dosage justification that refers to a version of the investigator's brochure other than the one filed;
  • data protection arrangements described in the protocol but absent from the information sheet;
  • annexes cited in the body of the protocol and missing from the dossier.

The timetable leaves little room

The reporting Member State validates the application within 10 days of its submission (Article 5). The Part I assessment report is submitted within 45 days of the validation date (Article 6); if the reporting Member State requests additional information, that period may be extended by up to 31 days. Part II follows the same 45 day period from validation (Article 7) and can likewise be extended. The single decision is notified within 5 days of the reporting date or of the last day of the Part II assessment, whichever is later (Article 8). An avoidable question from an assessor therefore costs more than two days of internal review.

Medical devices: do not carry these habits across

Regulation 536/2014 covers medicinal products for human use only. A clinical investigation of a medical device falls under Regulation 2017/745 and, in France, under Chapter V of the French Public Health Code, Articles L.1125-1 to L.1125-31: the ANSM carries out the scientific assessment and the personal protection committee (comité de protection des personnes) issues the ethical opinion. Performance studies for in vitro diagnostic medical devices fall under Regulation 2017/746 and Articles L.1126-1 and following. The reference good practice is ISO 14155, not ICH E6, and submission does not go through CTIS. A sponsor running both types of research is better off keeping two separate sets of templates and saying so in its procedures. You can submit your own templates and procedures so that a review applies them alongside the public texts.

In practice

Before filing, freeze the versions, then read in this order: the protocol against Annex I section D, the synopsis against the protocol, the investigator's brochure against the protocol, and the participant-facing documents against the Part II requirements. Give the cross-reading to someone who did not write the text. Keep a record of the discrepancies you corrected: that record is what will serve you at the first substantial modification. This is the work we support with EryonOne, whose covered frameworks include Regulation 536/2014: every finding cites the exact passage and the reference text, with a confidence score, and the decision to accept or set aside stays with the expert. The report is working material, not an approval.

Bibliographie