Assessment of US pathway for approving medical devices for rare conditionsBMJ 2014; 348 doi: https://doi.org/10.1136/bmj.g217 (Published 17 January 2014) Cite this as: BMJ 2014;348:g217
- Thomas J Hwang, researcher12,
- Daniel Carpenter, professor3,
- Aaron S Kesselheim, assistant professor2
- 1Faculty of Arts and Sciences, Harvard University, Cambridge, MA 02138, USA
- 2Program on Regulation, Therapeutics, and Law (PORTAL), Division of Pharmacoepidemiology and Pharmacoeconomics, Department of Medicine, Brigham and Women’s Hospital and Harvard Medical School, Boston, MA 02120, USA
- 3Centre for American Political Studies, Faculty of Arts and Sciences, Harvard University
- Correspondence to: A S Kesselheim
- Accepted 2 January 2014
In the US, manufacturers of high risk devices must submit data to the Food and Drug Administration (FDA) to demonstrate safety and effectiveness before the devices can be approved for wide use.1 The humanitarian device exemption (HDE) was established by the Safe Medical Devices Act of 1990 to encourage device manufacturers to develop products for treating or diagnosing rare diseases. Under the exemption, manufacturers do not have to provide the high quality data on effectiveness that would normally be expected for high risk devices, but they must still prove that the device is safe and that the probable benefits outweigh the risks from its use and that of alternative forms of treatment.
Policymakers in the US and Europe continue to examine and update the regulation of medical devices (box 1). Current EU proposals to require manufacturers of high risk devices to submit a clinical investigation report and publish a summary are similar to the requirements for HDE devices in the US. However, the EU proposals apply to all high risk devices not just those for rare diseases. We examine the scientific and regulatory characteristics of devices approved under the humanitarian device exemption since its inception and consider the benefits and risks of this pathway.
Box 1: EU medical device regulation
In Europe, responsibility for the approval of medical devices falls on commercial organisations known as notified bodies, which grant permission for manufacturers to display a Conformité Européenne (CE) mark if the device meets certain performance and safety criteria. Notified bodies are overseen by national regulators in each EU member state. Generally, a CE mark from one notified body allows the manufacturer to market the device in all member states.2
After controversies over the safety of devices, notably Poly Implant Prothèse (PIP) breast implants3 and metal on metal hip prostheses,4 5 the European Commission proposed revisions to medical device regulations in September 2012. The initial proposal aimed to strengthen the currently decentralised system of marketing authorisation and to improve traceability of devices in the supply chain.6 Some experts have argued that the proposed changes do not go far enough in ensuring that high risk devices are safe and effective.7 8 9
In April 2013 German MEP Dagmar Roth-Behrendt released draft amendments that would create a centralised approval pathway for innovative high risk devices, for which manufacturers would be required to present rigorous evidence of the device’s effectiveness and safety.10
Roth-Behrendt’s amendments were opposed by the device industry,11 12 which has argued that the cost and time burdens associated with gathering this evidence would reduce future investment in innovation.13
In October 2013 the EU parliament voted in favour of draft regulations that strengthen oversight of notified bodies but do not replace them with a centralised regulatory agency. Manufacturers would be required to submit a clinical investigation report and summary within one year of the end of the pivotal trial. Similar to the summaries published by the FDA, this documentation would be made available to healthcare professionals and the public.
The Council of the EU debated the rules in December 2013 but did not reach consensus. The proposed regulations may be amended again or reintroduced for voting before the close of the current legislative session on 30 June 2014.
Humanitarian device exemption
The FDA regulates medical devices according to their estimated level of risk to patients. High risk (or class III) devices, such as implantable pacemakers, receive the highest level of premarket scrutiny. The FDA stipulates that manufacturers submitting a premarket approval (PMA) application for a high risk device conduct “well controlled” clinical trials to prove effectiveness and safety.
For devices intended to treat rare diseases, manufacturers may alternatively seek to obtain marketing approval through the HDE pathway (table 1⇓). Rare diseases are defined as those affecting fewer than 4000 people in the US annually, such as dystrophic epidermolysis bullosa and complex ventricular septal defects that cannot be closed with standard surgical treatment. The pathway was established because the FDA recognized the need to adapt its oversight for rare diseases, given patients’ lack of therapeutic options and the difficulties manufacturers may have in conducting traditional trials to demonstrate effectiveness in rare diseases. In addition to the modified review standards, HDE applicants do not have to pay the standard user fees attached to regulatory review. After approval, HDE devices are subject to some restrictions—for example, physicians must obtain approval from the relevant local institutional review board before use.
The legislation has been changed since to 1990 to further stimulate therapeutic innovation for rare diseases. The original legislation stipulated that humanitarian use devices were not allowed to be sold for profit. This restriction was removed for devices labelled for use in children in 2007 and for all HDE devices in 2012.14
Requirements for approval
When conducting clinical trials for medical devices manufacturers usually first seek to establish proof of concept and safety in a small sample of patients—a pilot phase. The next stage is to establish data on effectiveness and safety from a larger number of patients.15 The FDA designates these as the “pivotal” trials used to make the approval decision (box 2).
Pivotal trials of HDE devices
The pivotal study (or studies) for each device is indicated by FDA reviewers in their summaries of data submitted by manufacturers. These studies provide the primary clinical evidence supporting the safety and probable benefit of the HDE device for the proposed indication.
Of the 55 devices approved under HDE with available data, five were approved without any pivotal trials: three referenced existing literature16 17 18 and the other two referenced comparable existing products (see appendix on bmj.com).19 20
Under the humanitarian exemption pathway manufacturers of devices must show reasonable assurance of safety and “probable benefit” from pivotal trials or other clinical investigations involving the device. The FDA also requires information on the size of the intended patient population, a statement that no comparable device exists, and a discussion of the risks and benefits of currently approved devices or alternative treatments for US patients. After a device is approved, the FDA publishes the approval order and a “summary of safety and probable benefit,” which gives the key data presented in the HDE application.
How has HDE been used?
We identified all devices approved through the HDE pathway through to September 2013.21 We characterised the types of devices, the medical specialty expected to use the device, and examined the evidence that underpinned the decision to authorise the devices (see appendix on bmj.com for full details of methods).
Since the start of the program, 57 HDE devices have been approved from 318 submitted applications (figure⇓). Of the 57 approvals, 40 (70%) were for implantable devices (table S1 on bmj.com). Most devices were intended to treat cardiovascular diseases (18), followed by orthopaedic (nine), and neurologic (eight) indications.
We also reviewed the postapproval requirements and subsequent regulatory action related to the devices. Twelve were approved with postmarketing study requirements, with a median minimum trial enrolment of 60 patients (interquartile range 32 to 122). Since the start of the program, five devices have been withdrawn, and the label for one device was revised after important safety issues were reported after authorisation (see below).
Among the 55 devices for which data were available (table 2⇓), 50 had trials that were classified as pivotal. Three devices had multiple pivotal trials, giving a total of 54 trials for analysis. We considered all other non-pivotal trials submitted to the FDA as supporting evidence.
Eight devices had been evaluated in randomised trials. Most of the preapproval pivotal studies (39) were single arm trials, and only one was fully blinded. Nine trials compared the device with an active control, such as the standard of care, but four of these used historical controls. For example, outcomes for the Berlin Heart EXCOR pediatric ventricular assist device were matched retrospectively with control patients from a registry of patients with mechanical circulatory support devices.22
The 54 pivotal trials enrolled a median of 38 patients (interquartile range 22 to 70). The supporting trials were similarly small: 18 devices submitted data from clinical investigations other than the pivotal studies, and these supporting studies enrolled a median of 79 patients (interquartile range 55 to 156).
Comparing humanitarian and standard devices
To provide insight into the flexibility offered by the humanitarian device exemption we compared the characteristics of trials underlying these approvals with those of devices approved through the usual more stringent PMA process. Specifically, we compared randomisation, blinding, comparator(s), and the size of the study with those described in a sample of premarket approvals of high risk cardiovascular devices by Dhruva et al23 and Chen et al24
Trials of humanitarian devices were less likely to be randomised (15% v 27%, P=0.046) and less likely to be double or single blinded (2% v 14%, P=0.005) than premarket approval trials (table 2⇑). Fewer HDE trials had an active control group (17% v 40%, P<0.001), and they enrolled significantly fewer patients on average (72 v 308, P<0.001).
Some in the device industry point to the advantages of speedier approvals of devices in Europe, where there is no requirement for effectiveness trials, than in the US.25 We therefore looked at whether devices approved in the US under the humanitarian device exemption came to market in the US before they had been approved in Europe, Japan, or Canada. We found that three quarters of HDE devices had received marketing approval from another regulatory body before receiving FDA approval. In particular, 38 (69%) had received EU approval at a median 2.5 years before FDA approval.
Performance of humanitarian devices
Some of the devices approved through the humanitarian exemption program may not have been available to US patients had this pathway not existed. For example, the vertical expandable prosthetic titanium rib is indicated for treatment of thoracic insufficiency syndrome in skeletally immature patients.26 Though subsequent follow-up studies suggested modest effect on lung function27 and a possible increased risk of ossification,28 the device has been valuable as a first step in spinal correction and improvement of quality of life for children with thoracic insufficiency syndrome and related disorders.29 30
Other devices approved through the HDE pathway have been controversial.31 For instance, one group of authors argued that manufacturers used the HDE approval for a deep brain stimulation device to bypass the rigorous premarket approval requirements and to market the device without a clinical trial.32 However, the device has been shown to produce clinically meaningful improvement, and it remains on the market as an adjunctive therapy for patients with severe, chronic, treatment resistant, obsessive compulsive disorder who have not responded to at least three selective serotonin reuptake inhibitors.33
Safety issues with some humanitarian devices have emerged after approval. A randomised controlled trial of an intracranial neurovascular stent34 led the FDA to restrict the label indication to a substantially smaller patient population after approval.35 Five devices have been withdrawn since the start of the program. Two devices indicated for recurrent cryptogenic stroke due to presumed paradoxical embolism through a patent foramen ovale in patients who have not responded to drug treatment were withdrawn in 2006 after the FDA ruled that the disease was not rare enough to meet the criteria for HDE approval.36 The reasons for three other withdrawals (two in 2010, one in 2007) were not stated.
Reflections and policy implications
Most HDE trials are small and lack key elements commonly associated with rigorous study design, such as randomisation, blinding, and active controls. Our analysis suggests that the HDE program allows FDA regulators to exercise substantial flexibility in approving devices with limited evidence in cases of important unmet medical needs. Since HDE devices are approved to treat rare diseases with few treatment options, single arm studies may have been the only ethical and feasible trial design. However, given the legal standard for approval, manufacturers may not have tried to conduct more rigorous clinical trials, even if such studies were technically achievable.
Clinicians and patients who need to make informed treatment decisions about devices approved through the humanitarian exemption should keep in mind the low level of evidence required for FDA approval and be aware that approval does not necessarily verify effectiveness. This may alter how patients and clinicians view the risks and benefits of such devices.
The FDA required postmarketing trials for less than a quarter of HDE devices, and mandated only small numbers of patients in those trials, despite the potentially larger pool of patients available after a HDE device is approved. We were not able to assess whether these postmarketing commitments were met.
We believe regulators should be more insistent about continued active study of HDE devices. Carefully controlled longer term observational studies of devices, or postapproval randomised controlled trials, would confirm or refute the safety and probable benefit identified in the small premarket studies. These would provide invaluable information on adverse events, malfunctions, and design deficiencies.37 38 In 2012, the FDA began developing a comprehensive postmarket surveillance system for medical devices.39 40 The agency should prioritise HDE devices in that system and consider stricter penalties for not complying with postapproval trial requirements.
Europe can also learn from the HDE system. The FDA publishes summaries of data provided by the manufacturers of approved devices, allowing others to analyse and discuss the evidence behind them. European device assessments are not publicly available, hindering scrutiny by independent researchers41 as well as by clinicians and patients using these devices. The lack of publicly available data on EU device approvals prevented us from further investigating the HDE devices that had received CE marking before FDA approval. Although EU approval does not guarantee that a device will be reimbursed by health systems or used by clinicians and their patients,42 43 the timing of approval suggests that these devices are being approved with even less clinical evidence than was presented to the FDA.
As the EU develops new regulations for medical devices, it is struggling with how to design a regime that protects patient safety while permitting efficient innovation. The current EU proposals for all high risk devices are similar to those for HDE devices in the US. Yet in the US even these devices are subject to further safeguard through additional oversight by the relevant institutional review board at the hospitals and healthcare facilities in which they are used.
Overall, when devising regulations, regulators should expect that manufacturers will generate the minimum premarket clinical evidence required to meet regulatory standards. In exceptional cases such as rare diseases the EU may want to adjust regulatory standards to reflect the level of unmet public health needs. It should also strengthen postapproval surveillance, such as by requiring postapproval trials, and penalties for their omission, or by mandating public registries enrolling a sufficient proportion of device users.
The humanitarian device exemption (HDE) program allows approval of devices for rare diseases based on proof of safety and probable benefit
Most HDE devices are approved on the basis of small, non-randomised, non-blinded, and non-controlled trials
HDE requirements are similar to those being proposed for all high risk devices in Europe
Postmarketing surveillance of HDE devices should be required more often given the low level of evidence submitted before approval
Cite this as: BMJ 2014;348:g217
Contributors and sources: DC and ASK have researched and written extensively on the laws and regulations affecting medical product approval and use in the US. This article was based, in part, on a previous study conducted by ASK on the clinical evidence underlying approval of drugs for rare diseases. TJH, while a student of DC and ASK, had the idea for the study of humanitarian device approvals. All authors designed the study. TJH undertook data collection, and TJH and ASK analysed the data and developed drafts of the manuscript. All authors contributed to interpretation of the results, drafting of the paper, and revisions of the manuscript. All authors contributed to and approved the final draft for publication. ASK is guarantor.
Funding: TJH received research grants for this study from the Interfaculty Initiative in Health Policy and Centre for American Political Studies, both at Harvard University. ASK was supported by a career development award from the Agency for Healthcare Research and Quality (K08HS18465-01), a Greenwall faculty scholarship in bioethics, and a Robert Wood Johnson Foundation investigator award in health policy research. The funders had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.
Competing interests: We have read and understood the BMJ policy on declaration of interests and have no relevant interests to declare.
Provenance and peer review: Not commissioned; externally peer reviewed.