When Coordinators Legislate: The HTA Coordination Group’s Guiding Principles on Data Transparency and the Limits of Institutional Self-Empowerment

1. The Legal Status of the Guiding Principles: Soft Law of Legal Effect

The critique that follows depends on a single doctrinal claim: the Guiding Principles are intended to produce legal effects, notwithstanding their formal disclaimer of binding force.

The reviewability test and its lineage. In Belgium v Commission (Case C-16/16 P, 2018, paragraphs 31–32), the Court of Justice held that ‘any provisions or measures adopted by the institutions of the European Union, whatever their form, that are intended to have binding legal effects’ may be challenged under Article 263 of the Treaty on the Functioning of the European Union (TFEU), the assessment turning on the act’s content, the context of its adoption, and the powers of the institution that adopted it. The phrase ‘whatever their form’ was chosen to defeat formal escape routes, summarising a consistent line of authority, from Grimaldi v Fonds des maladies professionnelles (Case C-322/88, 1989, paragraph 18) on recommendations, through France v Commission (Case C-325/91, 1993, paragraphs 22–23) on Commission communications, to Sogelma v EAR (Case T-411/06, 2008, paragraphs 36–37) on acts of EU bodies, confirming that EU instruments produce legal effects where their content imposes operational obligations, irrespective of formal label.

It is therefore necessary to set out the procedure within which the Guiding Principles operate. Joint Clinical Assessment confidentiality is decided in four steps under the HTA Regulation and Commission Implementing Regulation (EU) 2024/1381.

Step 1 Developer-led identification (Article 11(1) and 11(5) HTA Regulation; Article 19 of Implementing Regulation (EU) 2024/1381). The Health Technology Developer submits the Joint Clinical Assessment dossier under Article 11(1) HTA Regulation. On receipt of the draft reports, Article 11(5) HTA Regulation requires the developer to signal technical inaccuracies and to signal information it considers confidential and ‘justify its commercially sensitive nature’. Article 19 of Implementing Regulation 2024/1381 routes this correspondence through the HTA IT platform. The substantive identification originates with the developer, document by document, with reasons.

Step 2 Assessor and co-assessor processing (Article 11(3)–(6) HTA Regulation). The assessor and co-assessor, designated within the Joint Clinical Assessment Subgroup under Article 8(4) HTA Regulation, prepare the drafts, receive the developer’s Article 11(5) signals, and revise the drafts accordingly under Article 11(6) HTA Regulation. They are the first internal evaluators of confidentiality identifications.

Step 3 Joint Clinical Assessment Subgroup view and Coordination Group endorsement (Articles 12(1)–(2) HTA Regulation; Article 20(1) of Implementing Regulation 2024/1381). Article 12(2) HTA Regulation provides that the Coordination Group endorses the revised reports by consensus, with divergent opinions incorporated where consensus fails. Article 20(1) of Implementing Regulation 2024/1381 then provides that publication take place ‘after having considered the views of the Joint Clinical Assessment Subgroup as to the commercially sensitive nature’ of the information the developer has requested to be treated as confidential. This is where the Guiding Principles operate substantively: they have become the criteria the Subgroup applies in forming the views the European Commission (the Commission) must consider at step 4.

Step 4 European Commission procedural review and publication (Articles 12(3)–(4), 28(d), 30(3) HTA Regulation; Article 20 of Implementing Regulation 2024/1381). Article 12(3) HTA Regulation confines the Commission’s review to procedural compliance, not substantive re-assessment of confidentiality. Article 20(2) requires the Commission, before publishing, to inform the developer of the right to appeal, an appeal that runs back to the Commission itself, which serves under Article 28 HTA Regulation as Secretariat to the Coordination Group whose Subgroup gave the contested view.

It may be said in defence of the Guiding Principles that they promote transparency, that transparency strengthens confidence in Joint Clinical Assessment outcomes, supports national uptake of assessments, aligns HTA procedures with broader EU transparency trends (Alemanno, 2014; Dyrberg, 2003; Mendes, 2020), and reduces the risk that confidentiality claims hide clinically relevant evidence. On that view, the document does not create new obligations or determine what must be published; it simply helps the Commission and HTA participants apply existing legal rules consistently across Member States.

The issue, however, is not transparency itself but the rule’s legal basis. Even if the Guiding Principles reflect sound policy choices, the question is whether the HTA Coordination Group may set binding confidentiality criteria, and may set them through guiding principles rather than through the EU’s legislature using the EU’s ordinary procedure for adopting binding delegated legislation.

The Guiding Principles document defines the operative concept, declares categorical presumptions, and prescribes recourse (content); it is the substantive backbone of the Article 20(1) determination at step 3 (context); and the issuing body controls Joint Clinical Assessment procedure across steps 2–4, the Secretariat included (powers). All these facets show intention to have legal effects, and depart from the conferral of power in the HTA Regulation and Implementing Regulation 2024/1381..

A procedural objection nonetheless remains: step 4 places the publication decision with the Commission under Article 12(4) HTA Regulation and Article 20(1) of Implementing Regulation 2024/1381, so the Guiding Principles, issued by the Coordination Group, cannot have legal effects. The objection fails on four grounds. First, ‘whatever their form’ in paragraph 31 of Belgium v Commission exists precisely to prevent acts of legal effect from being insulated by a downstream decider; the test is operational, not formal.

Second, the relevant question is not ‘who signs?’ but ‘who determines?’ The Commission’s role at step 4 is operationally light by design. The Article 12(3) HTA Regulation review is procedural only; the Article 20(1) decision is taken by the Commission acting as Secretariat under Article 28 HTA Regulation, with no parallel substantive HTA expertise. Therefore, ‘having considered’ the Joint Clinical Assessment Subgroup’s views, in the ordinary course, means following them. Where the substantive determination is made upstream and the downstream signatory has neither the expertise nor the latitude to depart from it, the locus of legal effect is upstream, and the legal status of the upstream instrument is fixed by that effective determination, not by the formal signature.

Third, the cascade-access provision sits outside the four-step architecture: it governs who inside a Member State’s HTA organisation may view the dossier under national professional-secrecy obligations, with no Article 20(1) confidentiality decision interposed before access to the dossier is granted. Fourth, the conferred-powers problem migrates rather than is resolved: it moves from the Coordination Group to the Commission instead of disappearing. If the Coordination Group cannot lawfully adopt operative confidentiality criteria, the Commission cannot lawfully apply criteria not produced through any delegated legislation procedure. Either the criteria are the Coordination Group’s, in which case Article 3(7), points (c) and (h), HTA Regulation is the wrong basis, or they are the Commission’s, in which case comitology, as carefully calibrated in Articles 290 and 291 TFEU, has been bypassed.

The Guiding Principles therefore produce legal effects within the meaning of paragraphs 31–32 of Belgium v Commission. The rest of this essay proceeds on that footing.

2. Substantive Deficiencies

The definition of commercially confidential information is built without reference to, much less compliance with, the binding EU lex specialis on the subject, Directive (EU) 2016/943 on the protection of undisclosed know-how and business information, and without reference to Article 39.3 of the Agreement on Trade-Related Aspects of Intellectual Property Rights (TRIPS Agreement) (1994, as amended in 2017), which obliges World Trade Organization members to protect undisclosed test or other data submitted for marketing authorisation ‘against unfair commercial use’ and ‘against disclosure, except where necessary to protect the public’.

Article 2(1) of Directive (EU) 2016/943 defines a trade secret as information meeting three cumulative conditions: that it is ‘secret in the sense that it is not … generally known among or readily accessible to persons within the circles that normally deal with the kind of information in question’; that it ‘has commercial value because it is secret’; and that it ‘has been subject to reasonable steps under the circumstances, by the person lawfully in control of the information, to keep it secret’. Applied to a Joint Clinical Assessment dossier, substantial portions of submitted material may satisfy all three conditions. Indirect treatment comparison methodologies developed in-house, pre-specified subgroup definitions reflecting molecular or biomarker-driven enrichment hypotheses, sensitivity analyses disclosing the boundary conditions of comparative-effectiveness claims, and patient-level health-related quality-of-life patterns are not generally known to competitors; each derives commercial value precisely from its non-public status; and each is the subject of reasonable steps to keep it secret. The categorical declaration that ‘all clinical trial results and other clinical data’ and ‘evidence synthesis approaches taken including discussions on uncertainties’ are in general non-confidential operates as a blanket override of the Article 2(1) test in respect of the very classes of information for which the test is designed to grant protection.

Two further substantive concerns deserve mention. The asymmetry with the Joint Scientific Consultation regime is insufficiently justified: Joint Scientific Consultation outputs frequently reshape the pivotal trial design that becomes the Joint Clinical Assessment submission, yet they are fully confidential, while the Joint Clinical Assessment dossier they shaped is exposed. The cascade-access passage extends the HTA Regulation’s tightly drawn perimeter, designated Coordination Group and subgroup members, designated assessor and co-assessor, and patients and clinical experts under Article 11(4), to an undefined population of staff and external contractors. Although Articles 5(3), (5) and (6) HTA Regulation, with Commission Implementing Regulation (EU) 2024/2745, require all actors involved in joint work to file declarations of interest, the cascade-access construction does not clarify whether ‘other individuals within [the member’s] organisation’ are formally designated under that regime or accessed through a parallel national pathway. The population subject to HTA Regulation-level declarations therefore remains indeterminate from the developer’s perspective, unaccompanied by visibility or procedural recourse at Union level.

3. The institutional Problem

Beyond the substantive critique lies the structural one. The HTA Coordination Group has, in one instrument, defined a legal category, pre-classified entire bodies of dossier material against that category, retained for itself, through the Joint Clinical Assessment Subgroup, the discretion to apply that category under Article 20(1) of Implementing Regulation 2024/1381, and constrained the developer’s appeal to a procedure run by the Commission also acting as HTA Secretariat under Article 28 HTA Regulation. The HTA Regulation itself, in Article 11(5), provides that the developer ‘shall … signal any information it considers to be confidential and justify its commercially sensitive nature’, implying a developer-led, case-specific identification rather than displacement by an ex ante categorical declaration. By appropriating that interpretative power to itself, the HTA Coordination Group departs from the primary text.

This is incompatible with several first-order rules of EU public law. First, the principle of conferred powers (Article 5 of the Treaty on European Union (TEU); Garben and Klamert, 2026): Article 3(7), points (c) and (h), HTA Regulation empowers the HTA Coordination Group to ‘provide strategic direction’ and to ‘coordinate and approve the work of its subgroups’, not to adopt interpretative guidance with significant operational effects on confidentiality determinations. To find otherwise would be to stretch the meaning of ‘strategic direction’ and ‘coordinate’ too far. The choice of legal basis is itself anomalous: of the seventeen HTA Coordination Group documents adopted by 2 June 2026, the Guiding Principles is the only one anchored in those points of Article 3(7). If the Guiding Principles were presented as binding rules, they would require an Implementing Act under Article 291 TFEU or a Delegated Act under Article 290 TFEU.

Second, the Guiding Principles are incompatible with the Meroni doctrine (Case 9/56, [1958] ECR 133, at 152), as developed in Romano (Case 98/80, 1981, paragraph 20) and modernised in United Kingdom v Parliament and Council (the ESMA Short-Selling case, Case C-270/12, 2014). (Kohtamäki, 2026) The Court in C-270/12 accepted that an EU body could exercise substantial powers, but only because three conditions were satisfied there: the powers were expressly conferred by Regulation (EU) No 236/2012; the criteria for their exercise were precisely delineated in the legislative act; and the resulting decisions were amenable to judicial review (paragraph 53). Whether the Guiding Principles satisfy these ESMA safeguards is seriously open to question: no EU legislative act expressly confers on the Coordination Group the power to define ‘commercially confidential information’ or to declare categorical presumptions against confidentiality, and the criteria are not delineated by the legislator but written by the body that applies them. The Guiding Principles raise Meroni concerns about normative discretion exercised by a non-legislative body without Treaty-grounded delegation.

Third, the requirements of good administration (Craig, 2026) are breached. The concerns reflected in Article 41 of the Charter of Fundamental Rights of the European Union (the Charter), and the broader EU general principle of good administration, are engaged: a body that designs the substantive criterion, applies it, and adjudicates challenges to its application cannot satisfy the objective-impartiality requirement articulated in Ziegler v Commission (2013, paragraphs 154–155) and Chronopost v UFEX (2008, paragraphs 46–54). Article 298(1) TFEU reinforces the point at institutional level: ‘in carrying out their missions, the institutions, bodies, offices and agencies of the Union shall have the support of an open, efficient and independent European administration’. The HTA Coordination Group sits inside the EU institutional framework and is supported by the European Commission as HTA Secretariat under Article 28 HTA Regulation; the Article 298 standard applies to the architecture as a whole.

Fourth, the right to effective judicial protection (Biondi, 2026) (Article 47 of the Charter; Article 19 TEU) is not protected. An appeal under Article 20(2) of Implementing Regulation 2024/1381 to the Commission, which in practice takes its view from the Joint Clinical Assessment Subgroup, is not the independent review the Treaties require.

Fifth, the Guiding Principles may raise concerns regarding legal certainty. The Court of Justice of the European Union has consistently required that rules of law be clear, precise and predictable in their effects, so that those subject to them can plan their conduct accordingly. Health technology developers prepare Joint Clinical Assessment dossiers in reliance on the criteria fixed by the HTA Regulation and its Implementing Regulations, not on guidance issued or revised through instruments adopted under Article 3(7) of the HTA Regulation. To the extent that the Guiding Principles shift the operative criterion away from the case-by-case Article 11(5) regime toward an ex ante categorical regime, they impair the legal certainty on which dossier preparation depends.

EU regulatory and access-to-documents practice points in the opposite direction. Article 4 of Regulation (EC) No 1049/2001 admits exceptions to disclosure only on the basis of specific, concrete reasoning (Sweden v MyTravel and Commission, Case C-506/08 P, 2011, paragraph 76), and the concept of ‘business secrets or other confidential information’ under Article 339 TFEU is assessed by reference to case-specific criteria (Evonik Degussa v Commission, Case C-162/15 P, 2017, Grand Chamber, paragraphs 64 and 107). The General Court trilogy of 5 February 2018, PTC Therapeutics International v EMA (T-718/15, paragraphs 39 and 45), Pari Pharma v EMA (T-235/15, paragraphs 34, 36 and 38), and MSD Animal Health Innovation and Intervet International v EMA (T-729/15, paragraph 38), applied the same logic in the pharmaceutical-data context: no general presumption of confidentiality applies to clinical study reports, similarity and superiority reports, or toxicology reports under Regulation (EC) No 1049/2001, but the question whether specific information is commercially confidential must be assessed in concreto, document by document. Case-specific justification is the disciplinary anchor of the EU confidentiality regime in the regulatory sciences (see also EMA, 2014; EMA, 2019; and the broader trajectory of EU clinical-data transparency under Regulation (EU) No 536/2014, Recital 67). The Guiding Principles invert that discipline by categorical pre-classification.

4. Remedies

Before any contentious route is pursued, an amicable pathway should be considered. Article 29 HTA Regulation establishes a Stakeholder Network that includes health technology developers, patient and consumer organisations, non-governmental health-related organisations, and health professionals. The Coordination Group could refer the Guiding Principles back to that Network for structured consultation, with a defined timeline and a published response to the points raised. A revised text emerging from such a process would acquire the procedural legitimacy the present text lacks, would resolve several of the substantive concerns without the cost of contentious proceedings, and would itself demonstrate that the commitment in Article 5(1) HTA Regulation to act ‘in an independent, impartial and transparent manner’ extends to the production of guidance.

If that amicable route is not opened, or yields no meaningful revision, three pathways remain. First, if the Guiding Principles are recognised as having legal effects, an Article 263 TFEU annulment action brought by a privileged actor within two months of publication, on grounds of lack of competence, infringement of essential procedural requirements, and infringement of the Treaties and the Charter. Second, an Article 267 TFEU reference on validity in any individual Joint Clinical Assessment dossier dispute before a national court, attacking the Guiding Principles indirectly. Third, a complaint to the European Ombudsman under Article 228 TFEU for maladministration, focused on the absence of impact assessment, formal stakeholder consultation and Better Regulation procedures, and on the structural conflict between rulemaking and adjudication. The Ombudsman’s November 2025 finding that EU institutions must apply the rules they have established for themselves (European Ombudsman, 2025) provides an immediate doctrinal anchor.

5. Conclusion

The Guiding Principles are a small document with a large precedent attached. If the HTA Coordination Group can install operative confidentiality rules through ‘guiding principles’, the same route is open for PICO (population, intervention, comparator and outcome) methodology, for uncertainty handling, for subgroup analysis, for indirect-comparison standards. The HTA Coordination Group risks exercising functions analogous to rulemaking by accumulation, document by document, anchored each time in housekeeping competences, rather than in the delegated legislative procedures of Articles 290 and 291 TFEU.

Article 31 HTA Regulation provides for evaluation and reporting on the application of the Regulation, the first instalment of which is due no later than 13 January 2028. By that date, the HTA Coordination Group will have produced a cumulative methodological and procedural framework largely outside the legislative process, guidance on outcomes, on quantitative evidence synthesis, on multiplicity and on subgroup, sensitivity and post-hoc analyses, on validity of clinical studies, and now on data transparency. The question is therefore not whether one transparency text is constitutionally well-founded; it is whether the structural pattern of the HTA Coordination Group’s normative output has become a parallel normative framework before the first formal review of the Regulation is even due.

This framing shifts the issue from a transparency dispute to a broader governance question and points to the appropriate remedy: not revising the substance of a single document, but restoring the procedural boundaries the Treaties are meant to preserve.