Publication date: November 10, 2026
Clinical research, diagnostic testing and commercialisation under Polish and EU law
Law as at 5 November 2026
Abstract
A change in an epigenetic clock may justify further investigation of a medicine. Its value as evidence of benefit depends on what the clock measures, the population in which it has been validated and whether treatment effects on the score predict effects on health. The same assay may be used to analyse archived samples, select participants for treatment or sell a test to consumers. Each use raises a different legal question. This article traces those questions from trial design through to commercial use, with particular attention to Polish law on medical experiments, performance studies, access to medical records and the use of data from non-commercial clinical trials.
Keywords ageing biomarkers, geroprotectors, clinical trials, IVDR, biobanks, genetic data
Suppose researchers measure DNA methylation in volunteers’ blood before administering a substance and again six months later. In the intervention group, the mean score on the chosen clock falls by three years. The sponsor wants to describe the result as evidence that the product slows ageing. The first question concerns the comparator. Was the difference from placebo also three years, or does the figure describe only the change between two measurements in the same people? The second concerns the volunteers. Does a change of this size predict a later loss of independence, fewer hospital admissions or longer survival? Each claim calls for different evidence.[1]
A geroprotector, as used here, is a substance under investigation for its potential effect on biological processes associated with ageing. A model can be trained to estimate chronological age, mortality risk or the pace of physiological change. Two models that express their results in years can therefore be measuring different things. In an open-label pilot study in people with diabetic kidney disease, investigators reported fewer cells expressing markers of cellular senescence, including p16 and p21, in adipose tissue after dasatinib and quercetin. This provided evidence about a possible mechanism of these senolytic drugs. Establishing whether the intervention prolongs healthy life requires a study measuring that benefit or a reliable surrogate for it.[2]
The European Innovation Council recognised this problem in its Pathfinder challenge on Biotechnology for Healthy Ageing. For biomarker tools, the programme linked access to appropriate cohorts with the need to demonstrate a relationship between an ageing mechanism and clinical features. These conditions identify a research objective. Whether the results support a particular medicinal indication is a separate regulatory assessment.[3]
An ageing biomarker may identify a group at increased risk of functional decline. Such a prognostic marker makes events more likely to be observed during the study. A predictive marker helps identify patients likely to derive a greater effect from a particular treatment. A pharmacodynamic response marker serves another purpose. A change may show that the substance affects the biological pathway under investigation. If the sponsor uses that change to predict something that matters to patients, the marker is being proposed as a surrogate endpoint. The strength of that case depends on the biological rationale, the marker’s prognostic value and interventional evidence that treatment effects on the marker correspond to treatment effects on the clinical outcome.[4]
The CALERIE trial illustrates the distinction. In a post hoc analysis of samples from this randomised study of caloric restriction, the intervention slowed the pace of ageing measured by DunedinPACE. The authors found no significant change in the biological age estimates produced by PhenoAge and GrimAge. In a separate study, TranslAGE researchers applied 16 clocks to data from 51 interventional studies. Responsiveness varied with the intervention, population and length of follow-up. For a lawyer reviewing a regulatory dossier, the choice of clock is therefore material. Selecting the most favourable measure after inspecting the results weakens the inference, particularly when the investigators have made numerous comparisons. Whether the population studied resembles the population for the intended efficacy claim is a further question.[5]
A protocol should specify the algorithm version, tissue, sampling time, primary comparison and approach to multiplicity and missing measurements. A study of loss of independence also has to address what happens if a participant dies before the next functional assessment. An analysis confined to those who are well enough to attend the visit could give an overly favourable picture. ICH E9(R1) provides a way to link these decisions to the precise treatment effect the trial is intended to estimate. Choices made on this point can change the clinical meaning of the answer even if the assay is perfectly reproducible.[6]
Where a method is intended for use by several medicine developers, the European Medicines Agency (EMA) offers voluntary qualification within a defined context of use. Its Committee for Medicinal Products for Human Use (CHMP) may, for example, assess whether the tool is suitable for identifying a higher-risk population. Using the same tool as a primary endpoint requires a case for that additional purpose. Where the question concerns the design of one product development programme, the sponsor can seek scientific advice on the protocol.[7]
The biological nature of ageing does not by itself determine product classification or establish a proposed indication. Under EU and Polish law, a substance may be a medicinal product because it is presented as treating or preventing disease. It may also qualify because it can be used to restore, correct or modify physiological functions through pharmacological, immunological or metabolic action. In Hecht-Pharma, the Court of Justice held that the product must be assessed individually, taking account of factors including composition, dose and effects under normal conditions of use. Giving medicines legislation priority in a borderline case still requires a finding that the product meets the definition of a medicine.[8]
This distinction affects population selection. A study of whether treatment delays mobility loss in people at increased risk can define an event and a period of observation. A project in healthy adults that measures only a cellular marker will answer a narrower question. Experience with metformin in a different indication contributes knowledge about the substance. A new use still requires its own assessment of dose, duration, population and benefit–risk balance. A clinical trial authorisation allows the evidence to be generated. Bringing a new medicine to market or adding an indication to an existing authorisation requires a separate procedure. The authority considers, among other matters, whether therapeutic efficacy is substantiated and the benefit–risk balance is favourable for the proposed use.[9]
The interventions in the protocol then determine the legal route. Under Regulation (EU) No 536/2014 (the Clinical Trials Regulation, or CTR), a clinical study of a medicinal product constitutes a clinical trial if any one of the statutory criteria is satisfied. They include advance assignment to a therapeutic strategy outside normal practice, a prescribing decision taken together with the decision to enrol the participant or diagnostic or monitoring procedures additional to normal care. Observing treatment prescribed independently of enrolment can constitute a non-interventional study if the other criteria are also absent. Extra blood draws required by the protocol therefore merit scrutiny even if the medicine is already available from a pharmacy.[10]
A low-intervention trial is still a clinical trial. Apart from placebos, the investigational medicines must be authorised. Their use must conform to the authorisation or be supported by published scientific evidence of safety and efficacy in at least one Member State concerned. Additional procedures must pose no more than minimal additional risk or burden compared with normal clinical practice. An earlier stage of a programme may involve animal experiments. In Poland, these require authorisation by the competent local ethics committee. The assessment considers replacement of animals by other methods, reduction of animal numbers and refinement of procedures.[11]
Applications for medicinal product trials are submitted through the EU Clinical Trials Information System (CTIS). In a multinational trial, Part I of the assessment covers matters including the scientific rationale and benefit–risk balance. In Part II, each Member State concerned considers national matters including participant information, consent, recruitment and arrangements for compensation. The Polish decision is issued by the President of the Office for Registration of Medicinal Products, Medical Devices and Biocidal Products (URPL). The ethical assessment is carried out by the national bioethics committee or a designated committee on the statutory list.[12]
A healthy volunteer may have little prospect of a direct benefit from a preventive intervention. The application should explain what potential value to participants or public health justifies the foreseeable risks and inconvenience. In older populations, multimorbidity, concurrent medicines and the burden of study visits affect that assessment. EMA recommends the Short Physical Performance Battery to characterise physical frailty at baseline. It helps establish whom the study population represents. A proposal to use the same scale as the main measure of efficacy needs a justification appropriate to the indication.[13]
The English MET-PREVENT trial shows why that distinction matters. In older people with probable sarcopenia and physical prefrailty or frailty, metformin did not improve four-metre walking speed at four months compared with placebo. The authors also reported poor tolerability in this population. That finding concerns a particular dose, population and functional outcome. In another trial, a promising change in a laboratory marker would still have to be interpreted alongside tolerability, adverse events and the participant’s ability to function in daily life.[14]
Chronological age does not determine the ability to give informed consent. For participants who lack that capacity, Article 31 CTR requires, among other safeguards, consent from a legally designated representative, respect for the express wish of a participant capable of forming an opinion and a justification for studying that population. The sponsor must also plan safety oversight, reporting of suspected unexpected serious adverse reactions and publication of results with a lay summary, generally within one year of the end of the trial in all Member States concerned. Polish law addresses civil liability and insurance, with an exception to the sponsor’s insurance obligation for low-intervention trials. The Clinical Trials Compensation Fund provides a separate route to compensation, including for bodily injury or a health impairment resulting from trial participation.[15]
Regulation (EU) 2017/746 (the IVDR) governs devices intended to examine specimens from the human body, including software with the relevant intended purpose. The biological properties of the marker and the intended use of the tool that measures it are distinct issues. A retrospective analysis of blinded samples may examine whether changes in methylation correlate with subsequent functional decline. Where the result plays no part in participant management, the assay may have an exclusively research purpose. If the same result determines enrolment, assignment to a medicine or discontinuation of treatment, a decision about an individual now depends on it. Assessment must take account of the manufacturer’s intended purpose, instructions, promotional statements and actual role in the protocol. A “research use only” label cannot resolve the issue on its own.[16]
Consider the clock again. Dividing participants into subgroups for later statistical analysis has different implications from withholding the medicine from an individual whose score falls below a threshold. Companion diagnostic (CDx) status has a narrower test. The device must be essential for the safe and effective use of a corresponding medicine and identify people likely to benefit or those at increased risk of serious adverse reactions. As a rule, a companion diagnostic is a class C device under IVDR rule 3(f). In the conformity assessment, the notified body consults a competent medicines authority or EMA, as appropriate for the medicinal product.[17]
The performance evaluation addresses the scientific validity of the association between the measured feature and a clinical condition, analytical performance and clinical performance. An assay can measure methylation accurately yet assign patients to the wrong treatment group, for example if the threshold was derived in a younger population. Documentation should address the consequences of false positive and false negative results. The intended use covered by an existing CE mark must be checked against the proposed new purpose. Evidence that the assay performs as intended does not establish that the medicine works.[18]
Under the standard Polish procedure, a medicinal product trial and a device performance study requiring authorisation are assessed separately. URPL specifies an ethics opinion and authorisation by its President for a CDx performance study using samples collected specifically for that purpose. CDx studies using only left-over samples are subject to notification. The EU COMBINE pilot coordinates reviews of selected multinational combined studies. Poland is not listed among the participating countries in Annex I to the phase two sponsor guide. A sponsor therefore needs to plan the Polish national route even where sites elsewhere in the same project participate in the pilot.[19]
An ageing clock may be a statistical model, stand-alone medical software or a component of a diagnostic system. A program designed to interpret specimen results and provide medical information about an individual may fall under the IVDR. Software that draws on gait sensors and supports a diagnostic or therapeutic decision may instead fall under Regulation (EU) 2017/745 (the MDR). For a model that combines laboratory results, images and clinical history, the function of each module and the principal source of the information used for the decision matter. The manufacturer’s description should be tested against what users actually receive and what the product invites them to do.[20]
An application might provide general educational information on ageing. A different question arises if its report states that the customer’s profile warrants a particular therapy. Before deployment, the developer needs to establish the limits of validation, how uncertainty is communicated and how the algorithm performs across age groups and in people with multiple conditions. An update to the model or threshold may change how many patients are offered treatment. Preserving the trial version, recording changes and evaluating their effects makes it possible to reconstruct the basis for an individual decision.
If the software also meets the definition of an AI system, Article 6(1) of the AI Act requires a separate assessment. It covers an AI system that is a product, or a safety component of a product, subject to Union legislation listed in Annex I where third-party conformity assessment is required for that product. Regulation (EU) 2026/1744 moved the application of the high-risk requirements for this group to 2 August 2028. An exclusion for systems developed and put into service solely for scientific research and development requires reassessment when the system becomes part of a clinical offering. Medical device requirements apply on their own timetable.[21]
Polish laboratory medicine law regulates the activities of medical laboratories and the reporting of results to patients. Relevant duties concern staff qualifications, specimen identification, method validation, authorisation of results and quality standards. Genetic testing in healthcare may also engage Article 4 IVDR, including information about the nature and implications of testing and access to counselling for specified predispositions to conditions generally considered untreatable. If a hospital makes and uses its own assay, Article 5(5) IVDR calls for a separate assessment, including use within the health institution, no transfer of the device to another legal entity and an appropriate quality management system. Ordering a test online does not establish its regulatory status or the qualifications needed to interpret its result.[22]
The Polish Act on the Professions of Physician and Dentist treats research on biological material, including genetic material, taken from a person for scientific purposes as a medical experiment (Article 21(4)). Article 29a(2) deals separately with collection, processing, storage and distribution of material for scientific purposes. These biobanking activities are excluded from the definition. A biobank may thus preserve a sample under its procedures, while a subsequent project testing a specific hypothesis on that sample may require a favourable bioethics opinion, consent and insurance. Under this Polish regime a research experiment in a healthy or ill person may entail at most minimal risk proportionate to its potential scientific benefit. Article 23a excludes, among others, persons subject to a legal incapacitation order from such research experiments. These national conditions are distinct from the rules specific to medicinal product and device studies.[23]
Consider two plasma specimens stored in the same freezer. The first was drawn to establish an ageing research cohort and can be linked to its donor. The second is an anonymous remnant of routine diagnosis. In its TRANSMED guidance, the Polish Medical Research Agency distinguishes the original purpose of collection, the investigator’s access to the key and subsequent use of data in a clinical trial. These are practical pointers for assessing an individual project. Placing material in a biobank does not confer rights to carry out every future analysis. Article 28 of the Polish Act provides a particular route for scientific use of information from a medical experiment or screening where identification is impossible. That provision addresses information.[24]
Under Article 24(2)(7), participants in a medical experiment must be informed about foreseeable further use of results, data and material, including commercial use. For a sample bank, the practical decisions are whether new markers can be assayed, whether a specimen can be sent abroad, whether donors can be recontacted and whether an incidental finding with possible health implications will be returned. A research observation may first require clinical confirmation. The participant also needs a meaningful account of what withdrawal entails if material is still held, has been consumed in testing or has already contributed to aggregate results.[25]
In findings published on 22 October 2026, Poland’s Supreme Audit Office described genetic samples being sent to foreign laboratories without patients’ knowledge or consent. In a multicentre project, a material transfer agreement should identify the specimens and recipients, permitted analyses, conditions for onward transfer, return or destruction, and responsibility for breaching those limits. The parties must also establish the sender’s authority and check that the proposed transfer is consistent with what the donor was told. A signed agreement between institutions cannot supply permissions that neither institution holds.[26]
A clock score linked to a participant may reveal information about present or future health. DNA analysis can also entail processing genetic data. The controller must identify each purpose and the relevant lawful basis under Article 6 and condition under Article 9(2) GDPR. Reliance on the scientific research condition in Article 9(2)(j) requires a basis in Union or Member State law and safeguards under Article 89. Coding limits the visibility of a participant’s identity. Where reidentification remains reasonably possible, the data remain protected. This matters when rare variants are combined with age, clinical history or a small recruitment site.[27]
The European Data Protection Board distinguishes informed consent to trial participation from the legal basis for processing personal data. The sponsor’s duties to ensure safety and reliability serve a different purpose from scientific analysis. Withdrawal ends further interventions covered by the participant’s decision, but does not automatically remove data already lawfully obtained. Article 58 CTR requires archiving of the trial master file for at least 25 years after the end of the trial, subject to any longer Union requirements. This period does not determine how long every specimen is to be retained. Article 28(2) CTR allows a separate request for consent to use data outside the protocol exclusively for scientific purposes.[28]
Access to the source records raises a further issue. Article 26(4) of Poland’s Act on Patients’ Rights permits disclosure of medical records to a higher education institution or research institute for scientific purposes without revealing information that identifies an individual. A company working with the university needs its own basis for access. Consortium partners should identify actual controller and processor roles, access to the key and all recipients. Processing likely to entail a high risk requires a data protection impact assessment. A transfer to a recipient outside the European Economic Area engages Chapter V GDPR.[29]
The European Health Data Space (EHDS) will facilitate secondary access to data. Most of Chapter IV will apply from 26 March 2029. Article 51(1)(q) covers data from biobanks, while the date for genetic, epigenomic and genomic data under point (f) is 26 March 2031. An applicant will need to specify the categories and a permitted purpose. If a biobank holds both clinical observations and genomic data, the two elements of the requested dataset need to be classified separately. The Regulation provides for data permits, a secure processing environment, an opt-out and a ban on use for advertising. To obtain pseudonymised data, a researcher must justify why anonymous data would not achieve the purpose. In 2026, citing the EHDS provides no immediate entitlement to a Polish cohort or to physical specimens.[30]
A joint university and biotechnology project needs to establish what each party brings to the research and what rights it can acquire. A laboratory may control the assay technique, a university the model code and a hospital the records and specimens entrusted to it for a limited purpose. A licence to use an algorithm alone confers no right to train it on another cohort, disclose data to an investor or submit them to a medicines authority. Rights to material and data must therefore be distinguished from rights to software and inventions, and from publication arrangements. The allocation should also allow the records needed for regulatory verification to be retained and made available.
Articles 5 and 6 of the Polish Clinical Trials Act impose a particular constraint. Under the ordinary definition of a non-commercial trial, the sponsor and any co-sponsors must fall within the statutory categories, which include universities, healthcare entities and investigators. The right to dispose of the trial data must belong exclusively to those entities. Article 5(1), point 3, excludes from this route trials undertaken as part of the scientific advice specified there or as part of a paediatric investigation plan. Article 6(1) restricts use of data to secure a marketing authorisation, to vary an authorisation or for marketing. It makes an exception for variations concerning the safety of the medicinal product. Article 6(2) and (3) prohibit transactions by the sponsor that enable the uses set out there and make contrary transactions void by operation of law.[31]
Article 6(4) disapplies paragraphs 1–3 to trials within Article 5(2). Such trials must be funded entirely from public funds and meet at least the requirements relating to eligible sponsors and control of the data. The mere existence of a grant therefore calls for further enquiry into the full budget, partner contributions and contractual terms. Article 7 permits free or discounted supply of the study medicine and scientific, technical or financial support from a manufacturer or marketing authorisation holder, provided the support is disclosed in the application and the relevant authorisation is obtained. Disclosure of such support does not in itself give that partner a right to use the data for registration. This matters when an investor seeks an option over the results before enrolment begins.[32]
For patent purposes, the discovery of a biological relationship must be distinguished from its technical application. A reagent kit, a new method of analysing a sample ex vivo or a defined further medical use of a known substance may be assessed. Novelty, inventive step and industrial application matter. For diagnostic methods, much turns on how the claims are framed and whether they encompass the required steps of diagnosis practised on the body. Publication disclosing the invention before a filing can destroy novelty. Consortium terms should allow time for filing while preserving the obligation to make trial results public.[33]
A participant report, a scientific abstract and a retail advertisement may cite the same number, but they invite different inferences. A paper can report the change on a named clock, a confidence interval, the comparator and disagreement between prespecified measures. On a package or sales page, “reverses ageing by three years” will ordinarily be understood as a statement about what the product does to a person. A footnote linking to a paper still leaves the advertiser to establish that the evidence supports the claim and that the advertising rules for the product permit it.
Polish Pharmaceutical Law prohibits, among other things, advertising medicines that are not authorised in Poland, using claims inconsistent with the summary of product characteristics in medicines advertising, and advertising prescription-only medicines to the public. For a test, Article 7 IVDR prohibits attributing functions or performance the device does not possess or suggesting uses outside the purpose covered by the conformity assessment. The Polish Medical Devices Act also requires public advertising to be understandable to a layperson and restricts public advertising of devices intended for professional users. It therefore matters whether a test campaign is directed to laboratories, prescribers or consumers.[34]
Food supplements are subject to EU health claims rules. A reference to general, non-specific health benefits must be accompanied by an authorised specific claim under Article 10(3) of Regulation (EC) No 1924/2006. As a rule, food information must not attribute disease prevention, treatment or cure to food. Claims about reducing a disease risk factor have a separate authorisation route under Article 14. For a new ingredient, the questions include whether it falls within a category in Article 3(2)(a) of Regulation (EU) 2015/2283, whether it was consumed to a significant degree in the EU before 15 May 1997, whether novel food authorisation is required and which conditions apply to its listing. Polish notification to the Chief Sanitary Inspector on first marketing a supplement is a market monitoring measure. It does not establish an anti-ageing effect.[35]
The practical value of an ageing biomarker is that a clinical question can be investigated before a distant patient outcome can be observed. In the opening example, a first paper can accurately report the difference in the clock score against the control. Using that score to decide who receives treatment requires validation of the test and the applicable regulatory route. Claiming that the therapy prolongs independent living requires evidence connecting its effect on the score with precisely that benefit. These decisions can be made in sequence. A programme designed on that basis preserves the value of the first observation while making clear what must be demonstrated next.
KG LEGAL KIELTYKA GLADKOWSKI advises international life sciences and biotechnology businesses on Polish and EU regulatory requirements, research arrangements and commercialisation. Recognised by The Legal 500 for Healthcare and Life Sciences and Data Privacy and Data Protection, our firm combines sector knowledge with experience in corporate matters, negotiations and disputes.
We support projects where clinical research, technology and sensitive data intersect, helping clients address the legal requirements relevant to their research activities and business operations in Poland.
Contact: office@kg-legal.pl
#LifeSciencesLaw #Biotechnology #ClinicalResearch #DataProtection #PolishLaw #KGLegal
[1] ICH E9, Statistical Principles for Clinical Trials, section 2.2.6 (surrogate variables), EMA.
[2] L.J. Hickson et al., Senolytics decrease senescent cells in humans, EBioMedicine 47 (2019), pp. 446–456.
[3] European Innovation Council, EIC Work Programme 2026, section II.2.2, Biotechnology for Healthy Ageing, objective 2.
[4] ICH E9, Statistical Principles for Clinical Trials, section 2.2.6 (surrogate variables), EMA.
[5] R. Waziry et al., Effect of long-term caloric restriction on DNA methylation measures of biological aging in healthy adults from the CALERIE trial, Nature Aging 3 (2023), pp. 248–257. R. Sehgal et al., Responsiveness of epigenetic aging biomarkers to longevity interventions in humans, Nature Medicine 32 (2026), pp. 3477–3490.
[6] ICH E9(R1), Addendum on Estimands and Sensitivity Analysis in Clinical Trials, sections A.3–A.5, EMA.
[7] EMA, Qualification of novel methodologies for medicinal product development, procedural guidance for applicants, rev. 6 (2026). EMA, Scientific advice and protocol assistance.
[8] Directive 2001/83/EC on the Community code relating to medicinal products for human use, Articles 1(2) and 2(2). Polish Pharmaceutical Law Act of 6 September 2001, Article 2 point 32 and Article 3a, consolidated text, Journal of Laws 2026, item 612, as amended. CJEU, Hecht-Pharma, C-140/07, judgment of 15 January 2009, EU:C:2009:5, paras 29, 32 and 39–42.
[9] Directive 2001/83/EC, Articles 6, 8 and 26. EMA, Extensions of marketing authorisations: questions and answers, question 1.
[10] Regulation (EU) No 536/2014 on clinical trials on medicinal products for human use (CTR), Articles 1 and 2(2), points 1–4.
[11] CTR, Article 2(2), point 3. Directive 2010/63/EU on the protection of animals used for scientific purposes, Articles 4, 36 and 38. Polish Act of 15 January 2015 on the Protection of Animals Used for Scientific or Educational Purposes, Articles 3a and 42.
[12] CTR, Articles 4–8 and Annex I. Polish Act of 9 March 2023 on Clinical Trials of Medicinal Products for Human Use, Articles 9 and 21–25, consolidated text, Journal of Laws 2026, item 2.
[13] EMA/CHMP, Reflection paper on physical frailty: instruments for baseline characterisation of older populations in clinical trials, EMA/CHMP/778709/2015, 9 January 2018, sections 5–6. CTR, Articles 28–31 and Annex I.
[14] M.D. Witham et al., Metformin and physical performance in older people with probable sarcopenia and physical prefrailty or frailty in England (MET-PREVENT), The Lancet Healthy Longevity 6 (2025), 100695.
[15] CTR, Articles 31, 37, 41–43 and Annex V. Polish Clinical Trials Act, Articles 39–44.
[16] Regulation (EU) 2017/746 on in vitro diagnostic medical devices (IVDR), Articles 1(3)(a) and 2, points 2 and 12. MDCG 2022-10, Q&A on the interface between the CTR and the IVDR, May 2022, questions 5–7 and figure 1.
[17] IVDR, Article 2 point 7 and Article 48, Annex VIII, rule 3(f), and Annex IX, section 5.2. EMA, Companion diagnostics (in vitro diagnostics).
[18] IVDR, Articles 56–58 and Annex XIII. See also MDCG 2022-10, question 7.
[19] URPL, Performance studies of in vitro diagnostic medical devices (in Polish). See also the Polish Act of 7 April 2022 on Medical Devices, consolidated text, Journal of Laws 2024, item 1620, as amended. URPL, Questions and answers, questions 4–5 on combined studies (in Polish). European Commission, COMBINE Project 1, phase two sponsor guide, sections 1–2 and Annex I.
[20] MDCG 2019-11 rev. 1, Guidance on Qualification and Classification of Software, June 2025. Regulation (EU) 2017/745 (MDR), Article 2 point 1 and Annex VIII, rule 11. See also IVDR, Article 2 point 2.
[21] Regulation (EU) 2024/1689 (AI Act), Articles 2(6) and (8), 3(1) and 6(1), and Annex I. Regulation (EU) 2026/1744 (Digital Omnibus on AI), amendment to Article 113 of the AI Act, recital 38.
[22] Polish Act of 15 September 2022 on Laboratory Medicine, consolidated text, Journal of Laws 2025, item 1295, as amended. Polish Minister of Health Regulation of 30 June 2025 on laboratory quality standards, Journal of Laws 2025, item 961, Annex 3. IVDR, Article 4, in particular paragraphs 2–3. IVDR, Article 5(5), in particular points (a)–(c).
[23] Polish Act of 5 December 1996 on the Professions of Physician and Dentist, Articles 21, 23, 23a, 23c, 25, 29 and 29a, consolidated text, Journal of Laws 2026, item 37, as amended.
[24] Polish Medical Research Agency, TRANSMED I webinar answers (ABM/2024/8), p. 1, archived and coded specimens (in Polish). Polish Act on the Professions of Physician and Dentist, Article 24(2), points 5 and 7, and Articles 25 and 28.
[25] Polish Act on the Professions of Physician and Dentist, Article 24(2), points 5 and 7, and Articles 25 and 28.
[26] Polish Supreme Audit Office (NIK), findings on genetic testing and safeguards for genetic material, 22 October 2026 (in Polish).
[27] Regulation (EU) 2016/679 (GDPR), Article 4, points 5, 13 and 15, Articles 5, 6, 9 and 89, and recital 26.
[28] EDPB, Opinion 3/2019 on the interplay between the CTR and the GDPR, 23 January 2019, sections 2–3. CTR, Articles 28(2)–(3) and 58.
[29] Polish Act of 6 November 2008 on Patients’ Rights and the Commissioner for Patients’ Rights, Article 26(4), consolidated text, Journal of Laws 2024, item 581, as amended. GDPR, Articles 26, 28, 32 and 35, and Chapter V.
[30] Regulation (EU) 2025/327 on the European Health Data Space, Articles 51, 53–54, 66, 68, 71, 73 and 105.
[31] Polish Clinical Trials Act, Articles 5(1) and 6(1)–(3).
[32] Polish Clinical Trials Act, Articles 5(2), 6(4) and 7.
[33] Polish Industrial Property Act of 30 June 2000, Articles 24–29, consolidated text, Journal of Laws 2023, item 1170, as amended. European Patent Office, Guidelines for Examination 2026, G-II, 4.2.1.3, referring to G 1/04 and Article 53(c) of the European Patent Convention. European Patent Office, Guidelines for Examination 2026, G-VI, 6.1, Article 54(4)–(5) EPC.
[34] Polish Pharmaceutical Law, Articles 52–57, in particular Articles 56 and 57(1). Polish Medical Devices Act, Articles 54–55, read with Article 7 IVDR.
[35] Regulation (EC) No 1924/2006 on nutrition and health claims made on foods, Articles 2, 10 and 14. Regulation (EU) No 1169/2011 on food information to consumers, Article 7(3)–(4). Regulation (EU) 2015/2283 on novel foods, Articles 3, 6 and 10. Polish Food and Nutrition Safety Act of 25 August 2006, Article 29, consolidated text, Journal of Laws 2026, item 1399.