COMBINATION CORNER – Keys to a Robust Combination Product Design Verification & Validation

“How many samples do I need for testing?” This is the one of the most frequently asked questions as a device enters into the verification and validation phase of a device development program. Usually my response is one or all of the following:

-What does your Verification and Validation (V&V) Plan say about sample size, test method, and acceptance criteria?
-Have you conducted your risk assessment, and have the risk rankings been taken into account in establishing the acceptance criteria?
-Have you finalized your product requirements document (PRD)?

The responses to these questions shed light on whether or not the device development team has invested the appropriate amount of time to define critical design control activities that collectively would help identify the necessary samples required for testing. If these activities haven’t been completed, which is sometimes the case, the team simply isn’t ready to initiate design verification testing.

When discussing V&V testing, it is also important to keep in mind the difference between verification and validation. Verification means confirming the product has been designed correctly (robustness, reliability of performance against defined requirement/specifications, and integrity of product during worst-case conditions, such as temperature, drop, etc). Validation, on the other hand, seeks to confirm that the right product has been designed for the intended target market (is the final product what the end-user needs, and will it achieve the desired therapeutic benefit?).


The recent FDA regulations on Combination Products (CP) indicate that if the secondary constituent of the CP is a device, the device component development must be documented per, among other requirements, Design Controls. The Design Control process is typically defined as a multi-stage development approach from concept to commercialization. For the purposes of this article, the five phases are:

1. Concept
2. Planning
3. Design & Development
4. V&V
5. Manufacturing Transfer & Commercialization

In order to achieve a successful V&V (the fourth stage), certain activities within each of the prior phases needs to be thoroughly completed. It is within the context of this phased approach that two common mis-steps, which can hinder execution of a successful V&V, are further discussed.

1. Incomplete Requirements Within a PRD

In the planning phase of the project, the cross-functional team of drug + device team members will collectively develop the combination product requirements document (PRD). When the document is being drafted in the early phases of device and drug development, it is expected that there will be some requirements that are still unknowns that can only be established as either the device and/or drug is further developed. From a business and product development perspective, the team will need to determine whether it makes sense to delay device development efforts until those requirements can be defined or clarified.

Take as an example a new drug that will be delivered via a custom designed delivery device. The device will function as a container for the drug and diluent, upon activation mixes the two components, and then dispenses the final product topically onto the skin surface. The team collectively understands the type of device it wants to develop for its target customer base, yet the drug team at this stage of development is unable to confirm if the dose volume will be 3 mL or 6 mL. In the early planning stages, the PRD can be drafted to include a target range of volume, but the minimum dose volume requirement may need to be identified as a TBD. It is not uncommon for a development team to issue the PRD with a TBD, and during the Design Review, acknowledge that the TBD is present and needs to updated when additional information is available relative to the drug.

While the TBD was acknowledged in the planning phase, if the development team does not establish and commit to a timeframe by which the TBD must be resolved, the unresolved TBD will resurface as an issue when V&V protocols are being drafted or when testing is already completed and data analysis cannot be completed because the requirements have not been defined. While it seems like an unlikely scenario, these types of issues typically occur when teams fall victim to scope creep, tight timelines, critical team member turnover, or other factors.

It should also be noted that the TBD can even prevent the device team from proceeding to V&V because technically the Design Freeze (a critical product development milestone in the Design and Development stage) cannot happen without resolution of the open TBD requirement. Following on the example presented above, the drug volume is necessary to define corresponding device components for mixing, overall device size, or other design characteristics that are influenced by the necessity to accommodate the appropriate volume, but keep the device as small as possible.

2. Outputs of Risk Analysis are NOT Incorporated Into Test Acceptance Criteria

Conducting a risk assessment of the combination product is an activity that is initiated in the planning stages of the development process, and the assessment is continuously updated throughout the subsequent phases of the project as other development activities are conducted. It should be noted that the risk assessment is not a “one-and-done” document, rather it is a living document that continues to be updated throughout development and beyond product launch (commercialization).

During the assessment process, risks are identified, and mitigating factors are proposed to decrease identified risks to an acceptable level. The mitigating factors identify modifications/additions to the product requirements document or design specifications. In order to reduce the probability of occurrence, each risk mitigation must be evaluated to verify or validate that the risk has indeed been mitigated. The evidence of this evaluation is typically sourced from V&V test data. When developing the V&V plan, the test method and acceptance criteria must be identified for each requirement prior to conducting the V&V testing. If the risk assessment identifies that a specific device design feature carries a high risk of harm to the patient, the development team should consider this when establishing the acceptance criteria for the requirement. Typically, the greater the risk to the patient, the higher the parameters for the acceptance criteria as the team will want to have high confidence that the occurrence of that potential failure mode is minimal.

Let’s consider the example presented above in which we have a drug product that is delivered topically and another variation in which the product is delivered via subcutaneous injection. As part of the reconstitution and mixing process, the device design incorporates a custom needle that, upon user activation, pierces a rubber stopper and then transfers diluent from one chamber to another. The product requirement indicates that the reconstituted product shall have no particulate or coring present in the product delivered to the patient, and the test method for the requirement is identified as USP <381>, Fragmentation. Table 1 compares how the risk profile for each product application impacts the acceptance criteria, and ultimately the necessary sample size, for this specific requirement, while using the same device design.

The different applications (topical vs subcutaneous) directly influence the resulting risk present to the patient if particles are present in the final reconstituted product, and this is further translated into an acceptance criteria (or risk profile) that is tolerable to the development team for this specific requirement. It is not uncommon for companies to advocate that their products will be of highest quality standards. Yet when they are requested to provide a large number of samples for testing to align with these quality standards, they actively try to reduce sample size to the lowest possible value. In some instances, this may be appropriate, however, as demonstrated by the example above, if the resulting risk to patient is high, reducing the sample size is usually not the best approach.


In order to conduct a successful V&V on the intended combination product, the key is to understand that the V&V testing is not an isolated activity and task. Several of the activities identified in the Design Control process are key inputs toward establishing proper acceptance criteria and ultimately appropriate sample size for each requirement contained within the PRD.

The following is a checklist of items and/or reminders to keep in mind for combination product teams when preparing for V&V testing and that will also ensure compliance to Design Control.

If TBD’s Need to be Incorporated Into the PRD
– Assign responsibility to project team member and define time point when TBD should be resolved.

Document Responsibility & Timing in Design Review Meeting Minutes
– Resolve all design-related TBDs before Design Freeze milestone is completed.
– Ensure that TBDs are resolved before any formal V&V testing is initiated.

Leverage Risk Analysis Into V&V Activity
– Conduct a preliminary risk assessment of the device before testing is conducted on the device.
– Utilize outputs of risk assessment process as an input to defining acceptance criteria.
– Within V&V plan, establish acceptance criteria for each appropriate requirement prior to formal initiation of V&V testing.
– Update the risk assessment after testing is conducted on the device.

To view this issue and all back issues online, please visit

Lilli Zakarija is Co-Founder and President of EdgeOne Medical Inc, an ISO 13485-certified medical device testing firm and consultancy focused on supporting combination products through the device development (design control) process. Prior to founding EdgeOne Medical, she developed and led the global device engineering function for Baxter’s BioScience division (now Baxalta) in support of all its combination (biologic and device) products and single-use, disposable medical devices. Ms. Zakarija earned both her BS in Biomedical Engineering, and Masters in Engineering Management from Northwestern University, and her Executive MBA from Kellogg School of Management.