Frequently asked
Answers on clinical evaluation, clinical investigations, performance studies, evidence generation, and sponsor responsibility. Every question has its own shareable link.
As early as possible. Clinical strategy shapes the intended purpose and the claims, evidence requirements, study design, and regulatory pathway, and these decisions have downstream consequences across development, reimbursement, and market access.
Defining the clinical direction at an early stage avoids costly rework and ensures that evidence generation aligns with both regulatory expectations and business goals.
You do not design a clinical study without a State-of-the-Art (SOTA) review.
SOTA defines the current medical context: accepted standards of care, available alternatives, benchmark performance, known risks, and clinically relevant endpoints for the intended patient population.
Without it, there is no defensible basis for choosing comparators, setting acceptance criteria, defining endpoints, or justifying the risk-benefit profile, and the study will not withstand scrutiny from a Notified Body, ethics committee, or competent authority.
Under MDR and ISO 18969:2025, SOTA is the foundation that anchors the Clinical Evaluation Plan, the clinical investigation design, and the claims a manufacturer can credibly support.
Related engagementThe Catalyst
Under the EU Medical Device Regulation (MDR 2017/745), a clinical evaluation is a continuous process that assesses and analyses clinical data to verify safety, performance, and clinical benefit.
It should follow the methodology set out in ISO 18969:2025, the international standard for clinical evaluation of medical devices, which now serves as the reference framework previously provided by MEDDEV 2.7/1 Rev 4.
The output is a Clinical Evaluation Report (CER) supported by a documented Clinical Evaluation Plan, state-of-the-art review, and post-market clinical follow-up (PMCF) strategy.
Yes, but the more relevant questions are how much time you have, and whether the right competence is available to support the work.
Clinical evaluation, including the State-of-the-Art review, is a specific and structured process governed by MDR and ISO 18969:2025. In practice, it often takes more expert time to provide input and review on an internally drafted evaluation than to mandate an expert to conduct it from the outset.
Engaging an expert does not mean the manufacturer is relieved of work: the evaluation depends on product-specific knowledge, including inputs from the technical documentation, risk management file, and design history.
The process also surfaces strategic questions and decisions that are valuable in their own right, informing both product development and the commercial trajectory.
A clinical investigation (MDR Article 62) is a systematic study involving human subjects, designed to assess the safety or performance of a medical device, conducted in accordance with ISO 14155.
A clinical performance study (IVDR Article 58) evaluates the clinical performance of an in vitro diagnostic device using specimens from human subjects, and is conducted in accordance with ISO 20916.
Both serve to generate the clinical evidence required for CE marking, but the design, endpoints, and regulatory pathway differ significantly.
A proof-of-concept or proof-of-principle investigation is an early-stage, exploratory study designed to demonstrate that a device functions as intended in a clinical setting, to refine the design, and to generate preliminary safety and performance signals. It typically involves a small number of subjects, less rigid endpoints, and a flexible protocol intended to inform further development.
A pivotal clinical study, by contrast, is a confirmatory investigation conducted under a fixed and statistically powered design, with pre-defined primary endpoints, to generate the definitive clinical evidence supporting safety, performance, and clinical benefit for regulatory approval and CE marking.
The two serve distinct purposes within the clinical development pathway: the first de-risks the technology and informs design decisions, the second substantiates the claims that will accompany the device to market.
Both are required to demonstrate conformity under the EU In Vitro Diagnostic Regulation (IVDR 2017/746), but they address different aspects of device performance.
An analytical performance study evaluates the ability of the IVD to correctly detect or measure a particular analyte, addressing parameters such as analytical sensitivity and specificity, accuracy, precision, limit of detection, linearity, and interference. It is typically conducted using well-characterised specimens or reference materials, and does not require interpretation in a clinical context.
A clinical performance study (IVDR Article 58), conducted in accordance with ISO 20916, evaluates the ability of the device to yield results that correlate with a particular clinical condition or physiological state in the intended target population, addressing diagnostic sensitivity and specificity, predictive values, and clinical utility.
Analytical performance is a prerequisite: a device must perform reliably at the analytical level before its clinical performance can be meaningfully assessed. Together, with scientific validity, they form the three pillars of performance evidence required for CE marking under IVDR.
A PMCF study may be required when residual risks, long-term performance, or rare adverse events cannot be sufficiently addressed through pre-market clinical data alone.
Under MDR, PMCF is generally expected for most implantable and Class III devices, unless the manufacturer can provide sufficient justification that further PMCF activities are not necessary.
The PMCF Plan should be aligned with the Clinical Evaluation process and maintained as part of the Post-Market Surveillance (PMS) system, with regular updates based on emerging clinical data and post-market findings.
Clinical evidence should demonstrate that the device performs as intended, provides a meaningful clinical benefit, is clinically relevant, and supports conformity with the relevant General Safety and Performance Requirements (GSPR) under MDR Annex I.
The level and type of evidence expected depends on factors such as device classification, novelty, intended purpose, and the clinical and performance claims being made. Evidence may include literature data, post-market data, and clinical studies.
Strong clinical evidence is not only important for regulatory approval, but also for market trust, adoption, reimbursement discussions, and investor confidence. The goal is not simply to generate data, but to build evidence that supports both regulatory and commercial success.
Under MDR (Article 2(49)) and IVDR (Article 2(57)), the sponsor is the legal person who takes responsibility for the initiation, management, and financing of a clinical investigation or clinical performance study.
Sponsor responsibility cannot be transferred to an individual investigator, institution, or hospital: the manufacturer placing a device on the market remains accountable for the product, the evidence generated about it, and how that evidence is used in the regulatory file.
Investigator-initiated studies are a legitimate and valuable model, but they require the same rigour as manufacturer-sponsored studies when the data is intended to support CE marking, claims, or the Clinical Evaluation.
In practice, this means the study design, the Clinical Investigation Plan (CIP) or Clinical Performance Study Plan (CPSP), the choice of endpoints, and the statistical approach must be aligned with the intended purpose and the claims the manufacturer wishes to support, and must comply with ISO 14155 (MDR) or ISO 20916 (IVDR).
It also means the study must be authorised by the relevant Competent Authorities and ethics committees as a medical device or IVD study, not as an academic research project, and that data ownership, access, and transfer to the manufacturer must be agreed in writing before the study begins.
A common pitfall is for a manufacturer to provide a non-CE-marked prototype to a clinician who designs the study, runs it, analyses the data, and publishes the results independently. The intent is often genuine, but the consequences can be significant: endpoints that do not support the intended purpose or claims, documentation that does not meet MDR/IVDR expectations, missing regulatory authorisations, and data that sits with the institution rather than the manufacturer. The result is evidence that cannot be used in the technical documentation, and a study that has to be repeated.
The manufacturer's role is to remain the sponsor in substance as well as in name: defining the clinical strategy, ensuring regulatory compliance, and structuring the collaboration with the investigator so that the study generates evidence that is both scientifically meaningful and regulatorily usable.
It depends on internal capacity, complexity of the study, and the geographic scope. A CRO can bring scale, infrastructure, and regulatory expertise, but oversight, sponsor responsibility, and clinical strategy must remain with the manufacturer.
A further reason to engage a CRO, particularly for smaller medtech companies, is access to the core processes and quality systems required to run a clinical study to standard. Conducting a clinical investigation or performance study under ISO 14155 or ISO 20916 demands a defined set of processes, procedures, and tools covering study start-up, site management, monitoring, data management, safety reporting, vendor oversight, quality management, and archiving.
Smaller organisations rarely have all of these in place, and building them from scratch for a single study is neither efficient nor realistic. Delegating defined study activities to a CRO with mature, audited processes helps ensure that the work is performed correctly, that traceability and documentation meet regulatory expectations, and that the study remains inspection-ready throughout its lifecycle.
The decision should be based on a clear-eyed assessment of in-house competence, the processes and tools actually available, the level of support needed, and how the partnership will be structured and governed.
There is no single answer: timelines and budgets depend on the device, the study design, the indication, the number of subjects and sites, and the regulatory pathway.
As a broad indication, a focused proof-of-concept investigation at a single site may be completed within 6 to 12 months, while a multi-centre pivotal clinical study under MDR typically spans 18 to 36 months from finalisation of the Clinical Investigation Plan (CIP) under MDR, or Clinical Performance Study Plan (CPSP) under IVDR, to clinical study report, and longer when follow-up periods are extended.
Costs scale accordingly and may range from smaller early-stage investments to several hundred thousand euros for larger clinical investigations involving multiple sites, CRO support, and long-term follow-up.
The principal cost drivers are the complexity of the study design, sample size, number and geography of investigational sites, monitoring intensity, data management, and the duration of subject follow-up.
A realistic estimate requires a defined clinical strategy, a robust study design, and a clear view of the evidence the study is intended to generate.
Because the cost drivers in a clinical study sit in the detail, not in the headline.
A high-level synopsis typically describes the device, the population, and the broad objectives, but it leaves open the variables that actually determine effort and budget: endpoint definitions and how they are measured, inclusion and exclusion criteria, visit schedule and follow-up duration, sample size and statistical assumptions, monitoring strategy, data management requirements, safety reporting structure, and country and site selection.
Without these, any quote is a guess, and reputable CROs will either decline to price the work or attach such broad assumptions and caveats that the figure is not comparable across vendors.
To obtain a credible and comparable proposal, the manufacturer needs a robust study design, ideally captured in a draft Clinical Investigation Plan (CIP) or Clinical Performance Study Plan (CPSP), so that the CRO can scope the work against defined parameters rather than estimate against unknowns.