Using the Idea, Development, Exploration, Assessment, Long-term study Framework for Devices (IDEAL-D) to better understand the evolution of evidence surrounding fenestrated abdominal aortic endovascular grafts

Abstract

The use of fenestrated endovascular devices for repair of complex aortic aneurysms has increased to nearly 5,000 implantations annually in Medicare patients in the United States in recent years. Given that nearly all aspects of treatment for minimally invasive aortic intervention rely on medical devices to better care for patients with vascular disease, a clear understanding how new and innovative technology evolves over the life cycle of a medical device is an essential skill set for cardiovascular physicians.

Despite the need for this understanding, there is no standard framework upon which cardiovascular physicians, regulators, and patients can rely to better understand the evolution of evidence from product inception through adoption and long-term effectiveness evaluation. As the above devices are increasingly and broadly used, the need of a formal framework for regulation and device approval has emerged.

The goal of this review is to describe the Idea, Development, Exploration, Assessment, Long-term Study Framework for Devices (IDEAL-D). This framework is a model developed recently by an international panel of experts dedicated to better understanding the data steps necessary to bring a device from idea to routine practice to marketing, approval and monitoring. In this review, we use the example of fenestrated endovascular aortic devices to illustrate the IDEAL-D framework, and how it can help cardiovascular physicians improve their understanding of new technology, and the evidence which surrounds it from inception to long-term use.

Introduction:

Implantable medical devices, such as those used in infrarenal endovascular abdominal aortic endovascular repair (EVAR), have become an essential part of everyday cardiovascular care for patients with abdominal aortic aneurysms. The implantable devices which have enabled surgeons and interventionists to repair aortic aneurysms endoluminally have been studied for nearly 30 years¹, using a variety of observational, randomized, and registry based data sources^2–9.

Newer techniques in endovascular aortic surgery, such as fenestrated endovascular abdominal aortic aneurysm repair, have now become more common in vascular practice for patients with aortic aneurysms involving renal or mesenteric arteries. Increasingly complex endograft designs, with the incorporation of fenestrations and branches, have been used to treat juxtarenal, pararenal and thoracoabdominal aneurysms¹⁰.

During the initial evolution of this technology, single-center and limited multi-center studies demonstrated to vascular physicians that fenestrated endografts can be used to treat complex aneurysms, potentially offering lower mortality and morbidity rates compared to traditional open repair^11–14. Results collected worldwide from the use of fenestrated endografts have resulted in an increase of patient-customized devices, and industry remains focused on developing the ideal endograft.

However, little guidance exists for physicians, patients, and policymakers as to how this evidence should be studied and implemented when determining if this new technology is truly a step forward for patients facing high-risk treatments for cardiovascular disease¹⁵. Marketing approval for devices, both in the European Union (EU) as well as the United States, has historically focused on proof of safety and short term efficacy as a minimum requirement¹⁵. In the US, requirements for safety regulation requires are met with thirty day outcome data, and requirements for efficacy with one year outcomes. Once the above requirements have been attained, approval is then granted based on preclinical evidence alone, with no randomized control trials¹⁶. Fenestrated endograft devices often develop complications such as migration, stenosis, occlusion of visceral fenestrations and endoleaks beyond the one-year timepoint^11–14.

Currently in the EU, the CE mark requirements are being moved in a similar direction by the evolving medical device reform (MDR) program¹⁶. Similarly, in the US, more invasive devices now generally require a rigorous “pivotal” clinical trial either through the FDA’s pre-market approval (PMA) pathway or within the 510k pathway. Nonetheless, the pathway by which approval occurs, and by which evidence-based dissemination into practice is encouraged, is opaque to many who treat patients with vascular disease.

In this review, we discuss this pathway for evidence evaluation which emerged when a group of leaders in research, implantation science, clinical care, regulatory design, and evidence syntheses convened in 2009 to consider how devices evolve during the approval process. This framework, called the Idea, Development, Exploration, Assessment, Long-term study Framework for Devices (IDEAL-D), seeks to guide evidence evaluation at each step in a given technology’s life cycle. IDEAL-D’s potential to provide structure throughout a device lifecycle offers a multitude of benefits for physicians, patients and regulators alike, as they seek to best understand how to study and utilize new technology in the most efficient way possible.

Why a better framework for evidence development is necessary

Both EU and United States regulatory systems currently dichotomize device status as either pre-market (not yet approved) or post-market (approved). A system providing incentives for further evaluation and reporting at both early and later stages in the evolution of the device could create better evidence development and reporting and provide a more granular and descriptive pathway for evaluation at every “stage” of a given technology’s life cycle. If prospective registries were started from preclinical development used to monitor device performance through the outset of clinical use, late adverse events such as endoleak, endograft migration or branch occlusion would be captured early and addressed accordingly.

In the example of fenestrated aortic endografts, a framework such as this could potentially add value. It would help patients, physicians, and regulators better anticipate how this new technology will be studied over its life cycle, and what quality of evidence to expect when making decisions about the use of these new tools and techniques. Data requirements for device approval and surveillance can be matched to the device’s stage of development, known as total product life cycle evaluation ¹⁷. This framework is described in the review below to aid in dissemination for cardiovascular physicians, as was done with prior reviews published in Lancet,¹⁸ BMJ¹⁶, and the Journal of Vascular Surgery¹⁹. To add to this prior literature, this review uses the example of fenestrated endografts as an illustrative example for the framework.

The IDEAL-D framework

The IDEAL-D framework, derived from the IDEAL framework which describes the development of new surgical procedures, was created by an expert consensus group at a series of conferences and meetings to describe what types of studies and reporting should be used for new surgical devices, from first use through to adoption in practice¹⁷. The IDEAL framework consists of five main stages.

Stage 0: Preclinical

The first stage involves the preclinical development. For example, the application of the IDEAL-D framework to fenestrated endografts, would entail the in-vitro and bench testing involved in conceptualizing and testing devices before testing in-vivo. Early reports of device concepts have had to balance intellectual property concerns, and this stage necessitates careful cooperation between investigators and registry personnel. Product design, materials, and functional components were developed and tested to prepare the device for first-in-human (stage 1) studies. In general, stage 0 could include animal studies to optimize the technique before human studies begin¹⁷. Although reporting of preclinical research for fenestrated endograft devices are available, international minimum reporting standards for studies of therapeutic devices are lacking²⁰.

Stage 1: Idea, First-in-human

The current IDEAL-D calls for an international registry of all first-in-human studies with a description made available to the public. The registry enables the surveillance and identification of the benefit and harm reported by first-in-human studies. Possible device-related adverse events can be detected at an early stage, ensuring that unsuccessful innovation is not repeated through ignorance. Physicians and device engineers would take the opportunity to review the registry before embarking on a first-in-human study to avoid repeating a harmful error reported by another investigator.

For fenestrated endografts, an international registry was established in 2007 with the creation of the GLOBALSTAR project²¹. The goals of the above project included creating standardized follow-up protocols for patients undergoing endovascular repair of complex aortic aneurysms, developing reporting standards for the above techniques and providing an early warning system of complications specific to the techniques. The early results of the GLOBALSTAR registry were published in 2012 and included patients who underwent fenestrated endovascular repair of abdominal aortic aneurysms in the UK from 2007 to 2010²². Three year outcomes were reported as well. The GLOBALSTAR registry is in the process of recruiting and is a good paradigm of the importance of an international registry that will monitor, report and compare outcomes and complications of fenestrated endografts providing guidelines and indications for use to vascular surgeons all over the world. The ideal international registry will follow-up the recruited patients for a long-period of time identifying late complications such as endoleaks, migration, stenosis and occlusion of visceral fenestrations.

Stage 2: The Development and Exploration of New Treatment Paradigms

The second stage involves the development and exploration stages. The development stage is characterized by rapid iterative changes in the device and is reflective of learning curves and lessons experienced by the operator. In the exploration stage, the device is usually more widely used, and observational studies collect prospective data in order to facilitate learning and familiarity with the device.

In fenestrated endografts, this stage occurred in the first case reports which demonstrated safety, and potentially early efficacy of these new devices (Table 1). Further application of the IDEAL-D framework should describe the single center series as well as small, multicenter reports that demonstrate the effectiveness of the device in preventing aneurysm rupture and bolster the argument that these devices can be used in everyday practice.

Table 1 :

IDEAL Framework

Stage 0: The Idea	Stage 1: The Development of Evidence in Registries	Stage 2: The Exploration of New Treatment Paradigms	Stage 3: Assessing the New Treatment in Rigorous Ways	Stage 4: Evaluating Long-Term Outcomes
Question
Can the fenestrated device safely and effectively repair complex aortic aneurysms?	Which type of aneurysm anatomy would benefit the most? Any adverse events?	Is the device safe and effective in treating ruptured aneurysms?	Would a valid experimental study be required to further evaluate the device?	How well is the device doing compared to the previous established devices after long-term follow-up?
Aim
Proof of concept	Indications	Safety, efficacy	Efficacy	Comparative effectiveness
Patient base
Few	10s	100s	100s+	1000s
Optimal Study Design
In-vitro and bench testing of the device	First-in-human studies	Single center series or small multicenter reports	Randomized controlled trials, tracker trials or adaptive designs	Creation of prospective registries for safety surveillance

Stage 3: Assessing the New Treatment in Rigorous Ways

In the assessment stage, definitive observational studies should use a quasi-experimental study design with protocol driven controlled studies, standardized eligibility and prospective data collection. The question raised in this stage of the IDEAL-D framework is whether randomized controlled trials are necessary for novel devices that are similar to existing post-market devices. Even small modifications can potentially harm patients, suggesting that a clinical trial should be required for an implantable device^20,23. IDEAL-D allows for valid experimental designs, including randomized controlled trials, tracker trials, and adaptive designs. Studies based in economic modelling and reporting of health economic data for new devices are important additions to this stage^24–26.

Stage 3 reports related to fenestrated endograft would begin to compare this new technology with established treatment mechanisms. Specifically, this would involve observational or randomized studies comparing fenestrated treatments with traditional surgery. Many would argue that at present, fenestrated endograft is in stage III of this framework. A good example representing Stage 3 of the IDEAL framework is the UK-COMPASS Study, which is a risk-adjusted and anatomically stratified cohort comparison of open surgery, endovascular techniques, and medical management for juxtarenal aortic aneurysms (ISRCTN85731188). This study is planned to recruit patients from the spring of 2018 through June 2022 and is funded by the National Institute for Health Research and Health Technology Assessment Programme in England.

Stage 4: Evaluating Long-Term Outcomes

In this stage, established procedures are assessed for rare and long-term outcomes, and for variations in outcome. IDEAL-D recommends the design of registries for long term follow-up. Prospective registries can be used for safety surveillance^27,28. Individual surgeons can access the registries, share experiences and adverse events and identify device adverse events and structural inadequacies early on in the use of the new device.

Finally, for fenestrated endografts, long-term outcomes within registries would represent an invaluable source of information about the long-term clinical effectiveness of these devices in preventing rupture. The more granular the evidence collected in the registry, the better the resolution would be in discerning the factors associated with good and bad outcomes. For example, unusual presentation of endoleaks could be identified over time, and reporting of these events can prevent further complications such as a delayed rupture.

Discussion:

Highly innovative devices should normally be subjected to the entire integrated pathway of IDEAL-D stage studies, including prospective development and prospective exploration studies, randomized controlled trials, and registry studies. The extent to which these stages are operationalized in everyday practice has varied to date, and a better framework to guide physicians in undertaking these steps, and in evaluating new devices, may help improve this process.

Fenestrated endovascular grafts illustrate many of these challenges. The early evidence has largely evolved from single center studies, and large multicenter trials are lacking at present. There are several large, single center databases demonstrating the benefits of fenestrated endografts for juxtarenal and thoracoabdominal aneurysm repair [Table 2]. These nascent efforts in large-scale reporting of results^3,29–34 can serve as a starting point for future efforts. Further, while these devices have been studied in registries for short-term outcomes, few long-term, widely disseminated efforts at longitudinal surveillance in registries, have been accomplished. Therefore, there are both short and long-term gaps in the outcome assessment of these new devices, a challenge the IDEAL-D framework may help address.

Table II:

Outcomes of juxtarenal (JRAA) and thoracoabdominal (TAAA) aneurysm repair with fenestrated endografts.

Publication	Year	Patients	Aneurysm type	Perioperative Mortality (%)	Mean follow-up (months)	Branch patency
Verhoeven et al	2015	166	TAAA	7.8	29.2	94 (5-year)
Oderich et al	2016	127	JRAA, TAAA	0	9.2	96 primary 98 secondary (1 year)
Mastracci et al	2015	610	JRAA and type IV TAAA	1	92	NR
Ramanan et al	2016	134	JRAA	4	31	94
Sveinsson et al	2015	288	JRAA	2	11	NR
Eagleton et al	2016	354	TAAA (II and III)	4.8	36	97 (3-year)
Budtz-Lilly et al	2017	71	JRAA	2.8	36	92.7 (3 year)

Finally, randomized comparisons, either between device types or in comparison to open surgery, have not been accomplished. IDEAL-D may help to frame these challenges in important ways. For example, in new devices with incremental changes to a basic design, not seeking to substantially change the mechanisms of action but to compete on other grounds, the role of a randomized controlled trial would be controversial. If novel (first-in-kind) devices initiated a registry when they reach IDEAL-D stage 4, subsequent similar devices could be required to join. This might provide the infrastructure to conduct subsequent randomized trials comparing the new competitors with the first device, using the “trials within cohorts” or “nested within registry” design approach. This framework has been used in some instances already. Since 2011, when this collaboration started, the principles of IDEAL have been evident in the Food and Drug Administration guidance documents^35,36. Vallabhanemi et al published the nationwide early results of fenestrated endovascular repair in the United Kingdom²². In this study, 318 patients were collected from 14 centers through the GLOBALSTAR database. Perioperative mortality was 4.1% and intraoperative target vessel loss was 0.6%. Survival at three years was 89% and freedom from secondary intervention was 70%. Finally, investigators in the UK COMPASS trial have begun to enter patients in their registry, and these efforts will continue over the seven-year enrollment period of the study.

In summary, an IDEAL-D framework may help to guide the generation and collection of evidence for evolving innovations in aortic care. The case of fenestrated endografts represents a potential paradigm shift in regulatory evidence and may help improve the methods cardiovascular physicians use to develop, study, and implement new medical devices and technologies. We envision the creation of a prospective device database accessible to all hospitals, health care facilities and health care providers, independent from industry or third-party commercial groups. The IDEAL-D database will be self-funded by hospitals or federal and scientific funds. Data entry and maintenance of the database will require dedicated personel with clinical experience in the field of vascular surgery. Through a continuous evaluation process, more efficient ways to bring evidence to bear may “ideally” become achievable.

Figure 1 :

Stages in the IDEAL-D framework.