Why your combination product fails at clinical evaluation stage

Written by HATEM RABEH, MD, MSc Ing

Your Clinical Evaluation Expert And Partner

in
S

The regulatory framework for combination products sits at the intersection of two worlds. One governed by device rules. One governed by pharmaceutical rules. Most teams assume that if the device is the primary mode of action, they can treat the clinical evaluation as a pure device file with some drug safety data attached. This assumption creates problems that surface late in the review process.

\p>The MDR does not have a separate chapter for combination products. But Article 1(11) defines them clearly. A device incorporating a substance which, if used separately, would be considered a medicinal product, and which is liable to act on the body with action ancillary to that of the device. Article 117 then lays out the specific requirements. The medicinal substance must be assessed by a medicine agency or a notified body with pharmaceutical expertise. This is not optional.

What triggers the combination product pathway

The first mistake happens at classification. Teams look at their product and decide whether it is primarily a device or primarily a drug. They choose the regulatory pathway based on that decision. But the MDR does not ask which component is primary. It asks whether a substance liable to act on the body is incorporated into the device and whether that substance, if used separately, would fall under medicinal product legislation.

If the answer is yes, you have a combination product. Even if the device does 90% of the work. Even if the substance is well-known and widely used. The regulatory pathway is defined by the presence of the substance, not by its relative contribution.

The second mistake is assuming that a Certificate of Suitability or Pharmacopoeia monograph for the substance is enough. These documents confirm quality and manufacturing standards. They do not address the clinical performance of the substance in your specific device configuration, in your specific anatomical location, at your specific release profile.

The clinical evaluation must address both components separately and together

A clinical evaluation for a combination product cannot be structured like a device-only CER. You cannot write a section on the device, attach some drug safety data, and call it complete. The MDR requires you to demonstrate that you understand how the device and the substance interact, how each contributes to the clinical benefit, and how their combination affects the benefit-risk profile.

This means your clinical evaluation must include three distinct analyses. One for the device component. One for the substance component. One for the interaction between them.

The device component analysis follows the standard approach. You define the device characteristics, establish equivalence or generate clinical data, analyze the state of the art, assess benefit-risk, and plan post-market surveillance. This part is familiar territory.

The substance component analysis is where most files fall short. You need to address pharmacokinetics, pharmacodynamics, local and systemic exposure, dose justification, and potential interactions with patient medications. If the substance is released over time, you must justify the release profile and demonstrate that the dose remains effective and safe throughout the intended duration.

The interaction analysis is the most neglected. This is where you explain how the device modifies the substance’s behavior and how the substance modifies the device’s performance. Does the substance degrade the device material? Does the device coating affect drug release? Does the local tissue response to the device change the drug’s effectiveness? These questions must be answered with data or justified with scientific reasoning.

Why equivalence becomes complicated

Equivalence is already a difficult concept in device clinical evaluation. For combination products, it becomes even more restrictive. You can claim equivalence to another device. But you cannot claim equivalence to the substance unless your comparator product uses the same substance, at the same dose, with the same release profile, in the same anatomical environment.

I have seen manufacturers claim equivalence to a bare metal stent for the device component and equivalence to published drug studies for the substance component. Then they argue that the combination is equivalent by extension. This logic fails every time.

Equivalence requires that the combination product as a whole is equivalent to a comparator product. If no such comparator exists, you cannot use equivalence. You must generate clinical data specific to your combination.

This is a hard message to deliver to a project team that planned their development based on equivalence assumptions. But it is better to hear it early than to hear it from a Notified Body after you have submitted your technical file.

Article 117 consultation is not a formality

Article 117 requires that the quality and safety of the substance be verified by a competent authority for medicinal products or by a notified body with pharmaceutical expertise. This consultation is mandatory. It is not something you add at the end of the conformity assessment. It is part of the assessment itself.

The consultation must happen before the Notified Body issues your certificate. The authority or notified body with pharmaceutical expertise will review the substance’s quality data, manufacturing controls, and clinical safety profile. They will issue a scientific opinion. That opinion becomes part of your technical documentation.

If your Notified Body does not have in-house pharmaceutical expertise, they must involve another notified body or request consultation from a medicines agency. This adds time and coordination complexity to your conformity assessment process.

Many manufacturers discover this requirement late. They submit their technical file assuming their Notified Body will handle everything. Then they learn that a separate consultation is required, that it involves different timelines, and that the outcome could challenge their assumptions about dose, safety, or labeling.

Plan for this consultation at the beginning of your project. Not at submission.

The clinical investigation requirements are higher

If you need to conduct a clinical investigation for a combination product, expect more scrutiny. The ethics committee and the competent authority will evaluate both the device risks and the drug risks. The clinical investigation plan must address how you will monitor for adverse effects from both components. The informed consent must explain both types of risks to patients.

The investigator must have expertise in both device use and pharmacological management. This is not always easy to find. A cardiologist skilled in stent implantation may not have deep pharmacokinetics expertise. A pharmacologist may not perform interventional procedures. You need a team that can assess both dimensions.

Your clinical investigation protocol must include pharmacokinetic sampling if systemic exposure is possible. You must justify the dose based on preclinical data and any available human data. You must define stopping rules for both device-related and substance-related adverse events.

Regulators will ask why you chose that specific dose. Why that release profile. Why that duration of therapy. These are not device questions. These are drug development questions. Your answers must be scientifically robust.

Post-market surveillance must track both components

Your PMCF plan and your post-market surveillance system must be designed to detect issues related to the device, the substance, and their interaction. This means you cannot rely on device-only complaint categories. You need categories that capture drug-related adverse effects, unexpected pharmacological interactions, and issues that arise from the combination itself.

If your product is on the market and you receive a complaint about an allergic reaction, you must determine whether the reaction is to the device material, the substance, an excipient in the substance formulation, or a combination effect. Your vigilance system must be capable of that analysis.

Your PMCF plan should include targeted literature surveillance on both the device type and the substance. If new safety information emerges about the substance in other applications, you must evaluate whether it applies to your product. If new device complications are reported for similar devices without the substance, you must assess whether the substance changes the risk profile.

This is more work than a device-only post-market program. But it is what the regulation requires when you combine a device with a pharmacologically active substance.

How to structure the clinical evaluation report

The structure of your CER should make the combination nature of your product clear from the beginning. Do not hide the substance in a technical description section. State upfront that this is a combination product, cite Article 1(11) and Article 117, and explain how you will address both components in the evaluation.

I recommend a structure that separates the device analysis, the substance analysis, and the interaction analysis into distinct sections. Within each section, follow the standard CER logic: scope, clinical background, device description or substance description, data analysis, benefit-risk, and conclusions.

In the interaction section, address the following questions explicitly: How does the device affect substance release or distribution? How does the substance affect device performance or tissue integration? Are there potential interactions with patient medications? Are there subpopulations where the interaction changes the risk profile?

Your state of the art analysis must cover both fields. What is the current standard of care for the device type? What is the current standard of care for the condition being treated pharmacologically? Where does your combination sit relative to both standards?

This makes your CER longer. But it also makes your regulatory position defensible. When the Notified Body asks how you addressed the substance component, you can point to a dedicated section with a complete analysis. When they ask about interactions, you have a section for that too.

Why this matters beyond the technical file

Getting the clinical evaluation right for a combination product is not just about satisfying the Notified Body. It is about demonstrating that you understand what you are putting into patients. A device with a pharmacologically active substance carries risks that extend beyond mechanical failure or biocompatibility. It carries pharmacological risks, dose-related risks, interaction risks.

If you do not evaluate these risks systematically, you will not detect them in post-market surveillance. If you do not understand the interaction between device and substance, you cannot optimize the design. If you do not justify the dose scientifically, you cannot defend it when a competitor challenges your claims or when a safety signal emerges.

The clinical evaluation is where you prove to yourself, to the Notified Body, and eventually to clinicians and patients, that the combination is rational, safe, and effective. If that evaluation is incomplete, everything downstream is built on weak foundations.

I have worked on combination product submissions where the clinical evaluation was done correctly from the start. The review process was still rigorous, but it was predictable. The questions from the Notified Body were specific and answerable. The consultation under Article 117 was smooth because the substance had been properly characterized and justified.

I have also worked on submissions where the combination nature was underestimated. Those projects faced delays, repeated deficiency letters, and expensive protocol amendments. Some required new preclinical studies because the original data did not address the substance’s behavior in the device. Others required clinical investigations that were not originally planned.

The difference was not the complexity of the product. It was the quality of the clinical evaluation planning.

If you are developing a combination product, treat the clinical evaluation as a dual challenge from day one. Involve expertise in both device assessment and pharmacology. Plan for the Article 117 consultation early. Structure your CER to address device, substance, and interaction systematically. And be prepared to generate new clinical data if equivalence cannot be demonstrated.

The MDR does not give combination products an easier path. It gives them a more complex one. Recognize that complexity and plan accordingly.

Because the alternative is not a faster approval. The alternative is a major non-conformity and a realization that your clinical evaluation strategy was never aligned with the regulation.

Peace,
Hatem
Clinical Evaluation Expert for Medical Devices
Follow me for more insights and practical advice.