Translational Pathway for Transcatheter Aortic Valves
Preclinical Evaluation for TAVR
Carol E. Eberhardt
The term “preclinical testing” can be somewhat ambiguous. Within industry, the phrase ‘preclinical testing’ is often taken to mean ‘device in vivo/animal testing’, which is a final step in the product development process, described in more detail later in this section. But, for our purposes, this entire section will focus on the totality of development activities, including design verification and preliminary design validation activities; all of which must take place prior to performing human clinical trials, or more accurately, pre-human-clinical activities.
Quality Management System
Any organization that undertakes to manufacture and distribute a medical device must first develop a quality management system (QMS). Although a QMS can be structured in many forms to meet the needs of the organization, International Organization for Standardization (ISO) 13485 offers wide-ranging guidelines encompassing topics as broad as documentation practices and management responsibilities through product realization and measurement/analysis requirements. (ISO standards can be found at https://www.iso.org/standards.html.) Design and development activities such as planning and understanding design inputs and the processes that relate design inputs to outputs, as well as design verification and validation, are all key elements of a QMS. Process monitoring and measurement are critical to the day-to-day manufacturing of a consistent product as well as to identify boundaries for what constitutes an acceptable product. These limits must be identified and controlled through manufacturing process validation(s) and subsequent monitoring. A corrective and preventative action (CAPA) process is then implemented to deal with situations outside of process limits.
Design verification testing and validation activities, such as preclinical studies, are targeted at rigorous evaluation of the product, that is representative of an acceptable range of process results, in order to demonstrate that product and performance requirements are satisfied. One example of this rigorous evaluation is high cycle fatigue evaluation of the heart valve structural components. This design verification test is conducted on product representative of the highest stressed device configuration based on worst case dimensions and forces resulting from anatomical interaction superimposed with hypertensive blood pressures. The resultant stress conditions are typically multiplied by an appropriate safety factor for fatigue testing in order to ensure a margin of safety in the structural integrity of the device. Such considerations are basic requirements of a QMS and ISO 5840, and additional examples of such requirements will follow.
Initiation of a TAVR design development process should include early consideration of the risk profile for the intended patient population. Early focus in TAVR development should be directed at assessing risk and identifying risk mitigation strategies. For example, an appropriate risk assessment strategy for EFS designed to conduct FIH evaluations may be a descriptive risk analysis focused on identifying the information needed to address significant safety concerns and support early validation of basic device functionality. As data are obtained from these early implants, they should be applied to the development of a more thorough and quantitative risk assessment such as a failure modes and effect analysis prior to a pivotal clinical study.
One of the first considerations when developing a transcatheter aortic valve replacement (TAVR) device is to understand the design requirements based on the use conditions. This includes a system-level risk assessment that should be initiated early in the design development phase. Design requirements for TAVR systems can be roughly categorized as fitting into three fundamental areas: deliverability; structural integrity; and functional performance. Design verification testing and preclinical (in vivo) testing are both conducted to demonstrate that the design outputs (the list of specification blueprints describing how each device is manufactured) meet these design inputs, providing some measure of safety for further clinical evaluation.
Selecting the right tests based on the design and use conditions is crucial to ensure that the testing performed addresses all pertinent questions to sufficiently mitigate risk. Following a few basic steps can help in developing an appropriate test strategy:
— Asking the right questions;
— Clearly understanding the use cases and use conditions;
— Developing the right test methods; and
— Verifying that test methods appropriately challenge the design.
For each of these requirements, individual test protocols and reports are provided for each bench and laboratory test, computer modeling analysis (e.g., finite element analysis), and in vivo animal study. Each test report should include the purpose, test method, sample selection, sample size justification, test results, discussion of the acceptability of the results, conclusions, and when appropriate, justification and clinical applicability of the acceptance criteria.
ISO 5840-3:2013 provides detailed guidance regarding TAVR delivery system requirements. A new entrant to the TAVR market should apply the risk-based approach proposed by ISO 5840-3 and agree, through Pre-Submission meetings with the regulatory body, such as the British Standards Institution (BSI) for the European Union or the Food and Drug Administration (FDA) for the United States on the (delivery) system requirements to be assessed or justified (e.g., through comparison with reference systems with known clinical outcomes) in order to initiate a first-in-human (FIH), early feasibility (EFS), Conformité Européenne (CE), or investigational device exemption (IDE) study.
As defined by ISO 5840-3:2013, the design attributes to meet the intended performance of the delivery system shall take into account at least the following:
- the ability to permit consistent, accurate, and safe access, delivery, placement, and deployment of the transcatheter heart valve (THV) substitute to the intended implant site;
- the ability to permit consistent and safe withdrawal of the delivery system prior to and after deployment of THV substitute;
- the ability to minimize hemolysis;
- the ability to minimize thrombus formation;
- the ability to minimize blood loss (hemostasis);
- the ability to retrieve, reposition, and/or remove the THV substitute (if applicable);
- the ability to maintain coating integrity (if applicable); and
- biocompatibility of all tissue or blood contacting materials.
The design attributes to meet the intended performance of the transcatheter heart valve system should take into account at least the following:
- the compliance of the THV system with the requirements of ISO 10993-1 and appropriate other parts of ISO 10993;
- the visibility of the THV system under fluoroscopy or other imaging modalities;
- compatibility with magnetic resonance imaging (MRI); and
- the ability of the THV system to maintain its functionality and sterility for its specified shelf life prior to implantation.
The following excerpt from ISO 5840-3:2013 describes the testing of candidate device designs, and the attributes of the design that must be evaluated. Either single components, subassemblies, or final device assemblies are used as test articles in the testing suite, as justified by the manufacturer. Appropriate device conditioning per use conditions – for example, use of product that is fully representative of final manufacturing process or exposure to maximum sterilization cycles – is needed to ensure the integrity of the test results, and must be appropriately documented within the test report.
Implant interactions with delivery system
The manufacturer shall evaluate interactions between the implant and delivery system during use in accordance with the instructions for use (IFU) to ensure no damage is induced to the implant or delivery system. The following aspects shall be evaluated as applicable:
- crimping/loading and attachment of the device to the delivery system;
- loading device into the delivery sheath;
- positioning/deployment of the device within the target implant site;
- repositioning/recapturing of the device (if applicable) including damage to the valve if intended for immediate re-use;
- withdrawal of the delivery system from the patient;
- component dimensional compatibility with ancillary devices.
Loading of the device into the delivery system
The manufacturer shall define all specific performance parameters to be evaluated to verify safe and reliable loading of the device into the delivery system. The manufacturer shall demonstrate that the implantable device can be reliably attached to the delivery system in accordance with the IFU and satisfy attachment performance requirements, such as:
- attachment strength between the device and the delivery system;
- no damage to the device or the delivery system;
- crimped diameter;
- crimped shape (uniform or nonuniform);
- proper orientation of the device into the delivery system;
- dislodgement force;
- device sterility;
- device rinsing;
- delivery system flushing (de-airing);
- component dimensional compatibility with ancillary devices.
Ability to access and deploy
The manufacturer shall demonstrate that the attachment between the device and the delivery system shall be sufficient to permit safe, repeatable, and reliable delivery of the device to the intended implant site, release of the device from the delivery system, and safe removal of the delivery system from the patient in accordance with the IFU. The manufacturer shall define all specific performance parameters to be evaluated to verify safe and reliable deployment of the device within the intended implant site, such as:
- force to deploy;
- all relevant forces required to reposition the device (if applicable);
- flex/kink resistance;
- bond strength (tensile and torque);
- access angle between apex and annular plane for transapical delivery approach;
- time to deploy, including time of flow restriction or blockage, and time to restore flow;
- component dimensional compatibility with ancillary devices;
- balloon characteristics (if applicable);
- inflation/deflation time;
- relationship between the implant diameter and balloon inflation pressure, including assessment of effects associated with over-inflation and under-inflation;
- mean burst pressure;
- rated burst pressure;
- rated fatigue.
As transcatheter valve treatment therapy has evolved, and as competitive pressures have begun to play a role in device feature configurations, the following major system requirements are identified as necessary for the systems to be competitive in the current TAVR market:
- A low delivery catheter profile: <18-F outer diameter
- Atraumatic tracking: optimum tracking and flexibility through the vasculature
- Cost effective
- Re-capturability and re-positionability, for self-expanding frame-based technologies
- Ability to orient the rotational alignment with native commissures (clocking)
- Intuitive design (i.e., ease of use)
- Clear/smart position signaling
- Marker-bands/clear visual markers under fluoroscopy
- Stability during deployment
- Deployment time <1 min
- Efficient or minimized loading (into delivery system) process required for delivery of the device
- Material stability
- Shelf life >2 years
- Guide-wire compatibility
- Smart packaging (e.g., works with the delivery system for transcatheter aortic valve loading)
The frame, or supporting structural component, is a critical aspect of the development of a transcatheter-based valve heart valve. The frame must open the native valve without coronary artery obstruction, house the replacement valve, and withstand blood pressure loading and forces from interaction with the patient anatomy. Currently, success in TAVR procedures has encompassed both self-expanding and balloon-expandable frames. Though challenges vary from one to the other, procedural and design evaluations may be very similar.
The structural fatigue assessment process is comprised of multiple steps. Steps in the assessment might include:
- Characterization of use conditions for defining boundary condition inputs for modeling and simulation activities to assess applied stresses or strains;
- Determination of constitutive relationships for all materials included in the analysis;
- Validated stress analysis of the structural components to understand magnitude and location of peak stress/strain values resulting from imposed boundary conditions;
- Characterization of the material fatigue strength of the structural component(s);
- Fatigue safety factor or probability of fracture determination;
- Component level fatigue studies to supplement the material fatigue analysis.
An assessment of the corrosion resistance of all constituent metallic materials comprising the heart valve system is also a critical component of the structural integrity assessment. Appropriate corrosion studies may be performed as defined by risk assessment. These studies may include pitting / crevice corrosion, uniform / general corrosion, stress, galvanic, and / or fretting corrosion.
Several key areas must be considered with respect to functional performance of the TAVR device, beginning with basic tissue properties and ensuring compatibility with design and performance requirements. An aspect of the implantation procedure that can have an impact on tissue performance is crimping the TAVR into the delivery system and tracking the device through the anatomy. Crimp studies should be designed to understand any impact to the tissue integrity and tissue attachment to the frame.
It is also important that hydrodynamic performance and leaflet kinematics be characterized to ensure that the device provides acceptable flow performance to meet the needs of the target patient population. Any expected variation in valve shape that can impact flow performance should be well characterized and risks mitigated. As part of this evaluation, a thorough analysis of compromised leaflet kinematics, which might lead to early deterioration in vivo (i.e., thrombus formation, pannus, etc.), must be addressed with respect to any expected alteration of the deployed valve shape.
Additionally, an assessment of the TAVR durability should be performed to demonstrate adequate durability performance. TAVR durability assessments are typically performed through accelerated wear testing. In accordance with the requirements of ISO 5840-3:2013, transcatheter heart valve substitutes must remain functional for a minimum of 200 million cycles of accelerated wear testing. The durability test strategy should be comprehensive enough to determine the wear characteristics and anticipated failure mode(s) of the device.
Potential assessment strategies applicable for evaluating TAVR functional performance are summarized in Table 1.
Table 1. Functional Performance
|Impact of preservation techniques on tissue||
|Dynamic failure mode analysis||
These areas of device characterization and testing represent the basic needed understanding of device deliverability, structural integrity, and function over the expected device life. This successfully completed testing package is taken alongside preclinical animal (in vivo) testing, described in the following section, to prepare the researcher to move to consideration of human clinical testing. The bench top data as well as data retrieved from animal studies are expected to lower the risks of device use to arrive at an acceptable risk versus benefit profile to patients.
Design Validation (Preclinical) Studies
Per ISO 5840-3:2013, the overall objective of preclinical in vivo evaluation is to test the safety and function of the THV system in a biological environment with the closest practically feasible similarity to human conditions.
The preclinical in vivo evaluation represents the final investigational step prior to human implantation. Therefore, it should provide an appropriate level of assurance that the THV system will perform safely. Here again, a risk-based approach is utilized to determine what manner of in vivo animal testing is required. Introduction of new materials within the valve design often evokes the need for chronic timepoint testing. Designs whose materials have been already established as safe may require shorter-term or acute studies assessing deliverability, as applicable to the risks introduced by the attributes of the system being studied.
No single uniformly acceptable animal model has been established as the gold standard for long-term study of TAVR device safety and function. Therefore, the animal model(s) selected should be properly justified to ensure the highest degree of human-compatible conditions for the delivery system and test valve pertinent to the issues being investigated. Since chronic studies are conducted to elucidate heart valve substitute hemodynamic performance, biological responses, structural integrity and delivery system, and valve-related pathology in a specific anatomical position, it is preferable to undertake this longer-term testing of the valves in anatomical positions for which it is intended (orthotopic placement). Testing a valve in a position other than for which it is intended could produce false or misrepresentative results, as there are different flow conditions, pressures, and gradients across each of the four valves in the heart, all of which can affect the wear, fit, function, and calcification potential of the device over time. Therefore, much scrutiny and discussion of appropriate study endpoints is required for a non-orthotopically placed implant.
Orthotopic placement of a transcatheter aortic valve (TAV) in a preclinical model requires a deep understanding of the physical environment in which the TAV will reside. Although ovine and porcine models (most common preclinical models for TAV testing) overall have similar cardiac anatomy to humans, some components of the anatomy do differ. Specifically, humans have a much longer ascending aorta than either ovine or porcine models; therefore, the length of the ascending aorta, as well as the height of the sinotubular junction, must be investigated to ensure that the device has space to properly expand and allow flow both through the valve and retrograde into the coronary arteries. Excessive oversizing may take the valve out of round due to over-compression of the stent, creating regurgitation. Therefore, pre-implant screening (via computed tomography or echocardiography) is extremely important to ensure proper fit of the device within the annulus.
Aortic stenosis (stiffening and minimized motion due to calcification of the native aortic leaflets) is the disease state for which TAVs are indicated for treatment. Calcification is a process that occurs over many years in humans, and generally not until people reach 70 to 80 years of age. There is no established preclinical aortic calcification model in which to study TAV implants; as the calcification of the leaflets in humans provides some level of interference and anchoring of the TAV device. Additionally, unlike humans, who tend to have stable and flat left ventricular septal walls, many preclinical animal models have a sub-annular septal wall bulge that can interfere with the ventricular edge of the TAV after implant. Therefore, the lack of calcification and the presence of a prominent and often highly mobile septum in preclinical animal models should be considered when implanting the device, as there will likely be a greater propensity for the device to migrate out of a smooth annulus when a large muscular shelf interferes with it. Screening can help eliminate animals with highly mobile septal shelves, but that does not change the fact that the preclinical models used do not truly represent the disease state that the TAVs are indicated to treat. Therefore, migration resistance should not be tested in a preclinical model, or alternatively, migration resistance performance within the animal model should not be used as a predictor of expected migration performance when devices are used in humans. Additionally, preclinical testing cannot assess the impact of the TAV on the calcified nature of the native leaflets in humans.
Finally, study duration must be used to drive the choice of preclinical animal models. If using a juvenile or nonadult animal in a long-term study (>90 days), the growth rate of the animal needs to be taken into consideration, as the animal could outgrow the device implanted, affecting both fit and function of the device over time. In general, studies longer than 90 days should rely on adult animals, which are not expected to increase in size after a certain age as their growth curve is relatively flat. For example, it is well accepted that the preclinical model seen as the gold standard for assessing calcification in a surgical valve is to implant it in a juvenile sheep and conduct a 150-day study in the mitral position. However, this same model cannot be used for TAV implants because of poor device fit in a sheep model and the animal’s growth. Additionally, while a surgical valve is sewn into place, a TAV must rely on the surrounding anatomy to stay consistent over time. What is considered the gold standard with one technology cannot be assumed with another.
The concurrent implantation of reference heart valve substitutes enhances the comparative assessment by providing a bridge to known clinical performance. In addition, such an approach facilitates the distinction between the complications related to the reference heart valve substitute versus those of the THV system.
At some point, bench testing and animal studies don’t help refine the therapy enough to make it better. Iteration is a part of innovation, which emerges from engineering and continues until there is the necessary point of “design freeze.” This is where innovation leads to research and validation through some sort of trial.
The FDA has acknowledged that requirements for investigational device exemptions vary depending on the potential for device changes based on the clinical results. Toward this end, FIH and/or EFS trials have evolved to allow the manufacturer to obtain feedback on innovative new concepts and designs, in keeping with the risk-based approach in constructing the study pathway.
Information from these EFS studies would be expected to impact the system and device design(s) prior to initiation of the pivotal clinical trial. This allows the translational researcher to gather data early in the development process and apply that data to device improvements. A traditional feasibility study (TFS) may be included or conducted in lieu of an EFS near the end of the design development process or after design freeze. Since this study is conducted near the end of development activities, a higher level of preclinical or prior clinical data would be anticipated. Likewise, the ISO has transitioned from a requirement-based approach to a risk-based approach to enhance innovation in the cardiac medical device field.
As a general methodology, one means of justifying lesser cycle count (and thereby shortening test duration) would be to cycle a device at higher pressures or forces. This may be a realistic approach to early initiation of a FIH study.
It should be noted that there can be differing requirements between the FDA and the multiple European Union notified bodies. For devices that would be considered next-generation devices in which clinical data exists on a similar device, the tabulated values may be more realistic. For new devices, that is, those utilizing different materials and/or designs, the manufacturer should rely more heavily on a risk-based approach to identify hazards and means of mitigation. For example, a polymer leaflet-based device may have sufficient low cycle fatigue characteristics yet need to demonstrate assurance of high cycle fatigue to ensure full understanding of the potential failure modes. Furthermore, a thorough risk assessment may identify a need for mitigation of thrombus formation, such as surface modification. This, of course, will generate additional testing to prove efficacy of the surface treatment.
Upon successful completion of the design process and validation of the design via clinical study, the manufacturer must demonstrate the capability of producing the device repeatedly within the required specifications. This ensures that the mass-produced devices deliver the results observed during design validation; it can be a significant undertaking. Validation of the individual manufacturing processes requires an evaluation of those critical processes that drive the functional performance of the devices and demonstrate objectively that those processes are stable and show repeated conformance to specifications. In general, the activities are divided into installation, operational, and performance qualifications.
- Installation qualifications – Performed to document objective evidence that all equipment used within a process has been installed correctly and all functions are operating as intended.
- Operational qualifications – Performed to document evidence that a process, when performed at its extreme high/low parameters, produces products that meet the required process outputs. This evaluates the robustness of the process.
- Performance qualifications – Performed to document evidence that a process can repeatedly produce products that meet required outputs consistently over multiple lots/batches.
Manufacturing validations are performed as a requirement for commercial distribution of the device upon regulatory approval. These activities can often be equal in duration as the design process and require a great deal of planning and resources for execution. Additionally, these activities may not be completed when a device is being studied in human clinical trials, and a manufacturer may rely on verification of the outputs of each lot/batch produced, rather than investing time in the validation of a process that could change in response to validation data or further design changes. Some processes within overall manufacturing are identified as highly critical (e.g., sterilization) and may require continuous monitoring even after the process has been validated. Further, all manufacturing validation activities are also subject to the manufacturer’s QMS and must maintain an updated level of documentation.
The following standards (which can be found in more detail at the ISO website at https://www.iso.org/standards.html) are all relevant to provide guidance for the manufacturing process as described:
- ISO 5840-1:2015 – Cardiovascular implants – Cardiac valve prostheses
- ISO 5840-3:2013 – Cardiovascular implants – Cardiac valve prostheses – Part 3: Heart valve substitutes implanted by transcatheter techniques
- ISO 14971:2007 – Medical devices – Application of risk management to medical devices
- EN ISO 14971:2012 applies only to manufacturers with devices intended for the European market; for the rest of the world, ISO 14971:2007 remains the standard recommended for medical device risk management purposes
- ISO 13485:2016 – Medical devices – Quality management systems: Requirements for regulatory purposes
- ISO 10993-1:2018 – Biological evaluation of medical devices – Part 1: Evaluation and testing within a risk management process
- FDA Guidelines – Use of International Standard ISO 10993, “Biological Evaluation of Medical Devices Part 1: Evaluation and Testing” (1)
- U.S. Food and Drug Administration. Use of International Standard ISO 10993-1, “Biological evaluation of medical devices – Part 1: Evaluation and testing within a risk management process”. Guidance for Industry and Food and Drug Administration Staff. Available at: https://www.fda.gov/media/85865/download. Accessed May 6, 2019.