7
Medical Devices: Research and Development for Rare Diseases

The Vertical Expandable Prosthetic Titanium Rib (VEPTR), a device that has saved the lives of 300 infants and young children who otherwise would have died from lack of breath [thoracic insufficiency syndrome], has been approved by the U.S. Food and Drug Administration (FDA)… . The titanium rib is curved like a ribcage and has holes that allow the surgeons to expand the device in outpatient surgery every six months. The rib is implanted in infants as young as 6 months and in teenagers until skeletal maturity, typically age 14 in girls and age 16 in boys… . “It took 13 years to gain FDA approval because it took a long time to accumulate a lot of patients with rare diseases” Dr. [Robert] Campbell [the inventor] said.

UTHSCSA, 2004


For rare diseases, efforts to accelerate research and product development clearly focus on drugs and biological products. Devices and the need for devices are much less frequently mentioned in articles or conversations. When devices for rare conditions are discussed, it is generally in connection with pediatric populations.

To acknowledge the emphasis on drugs for rare diseases is not to imply that devices are not important for many people with rare medical conditions. Some people depend critically on devices targeted at distinctive features of their condition, for example, children who have received the implanted titanium rib described above. No pharmaceutical or biological product can provide the mechanical support afforded by this implant. Ge-



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 205
7 Medical Devices: Research and Development for Rare Diseases The Vertical Expandable Prosthetic Titanium Rib (VEPTR), a device that has saved the lives of 00 infants and young children who otherwise would have died from lack of breath [thoracic insufficiency syndrome], has been approved by the U.S. Food and Drug Administration (FDA). . . . The titanium rib is curved like a ribcage and has holes that allow the surgeons to expand the device in outpatient surgery every six months. The rib is implanted in in- fants as young as  months and in teenagers until skeletal maturity, typically age  in girls and age  in boys. . . . “It took  years to gain FDA approval because it took a long time to accumulate a lot of patients with rare diseases” Dr. [Robert] Campbell [the inventor] said. UTHSCSA, 2004 For rare diseases, efforts to accelerate research and product develop- ment clearly focus on drugs and biological products. Devices and the need for devices are much less frequently mentioned in articles or conversations. When devices for rare conditions are discussed, it is generally in connection with pediatric populations. To acknowledge the emphasis on drugs for rare diseases is not to imply that devices are not important for many people with rare medical conditions. Some people depend critically on devices targeted at distinctive features of their condition, for example, children who have received the implanted titanium rib described above. No pharmaceutical or biological product can provide the mechanical support afforded by this implant. Ge- 0

OCR for page 205
0 RARE DISEASES AND ORPHAN PRODUCTS netic tests that are necessary for the diagnosis and treatment of certain rare conditions are, in certain cases, regulated as medical devices. In addition, people with rare conditions benefit from a large number of medical devices that are used generally in connection with complex surgery, anesthesia, re- spiratory support, nonsurgical cardiac procedures, administration of certain medications, diagnostic and therapeutic imaging of various kinds, labora- tory testing, and other services. Clinical studies of the titanium rib were supported under the orphan products grants program described in Chapter 3. Earlier, the National Organization for Rare Disorders provided a seed grant from its donated research funds. The two companies that were involved in manufacturing the device for research use participated out of interest in children’s health rather than expectations of profit (Campbell, 2007). After years of inves- tigation and adaptation, the device was approved by the Food and Drug Administration (FDA) in 2004 through a Humanitarian Device Exemption (HDE). This process was established in the Safe Medical Devices Act of 1990 (P.L. 101-629) to provide incentives for the development of medical devices for small populations. Although medical devices for small popula- tions are grouped under the label orphan products in the grants program created by the Orphan Drug Act and are within the charge of the Office of Orphan Product Development (OOPD), the term orphan medical device does not appear in legislative or regulatory language. Regulatory requirements and product development pathways differ significantly for medical devices compared to drugs and biologics. Thus, this report devotes a separate chapter to medical device development, regu- lation, and reimbursement. This chapter begins with a brief overview of important differences between devices and drugs. It then reviews device regulation and reim- bursement with an emphasis on the HDE process and other policies or procedures that are potentially most relevant to complex, high-risk devices intended for small populations. This discussion is followed by an overview of the research and development process for complex devices and a discus- sion of barriers and opportunities for the development of devices for small populations. As this chapter highlights, the stringency of government regu- lation of devices is related to the risk presented by the device. DIFFERENCES BETWEEN MEDICAL DEVICES AND DRUGS Compared to pharmaceuticals, medical devices are an extremely di- verse group of products. Some are as simple as adhesive bandages, tongue depressors, and plastic tubing. Others are complex, for example, various implanted cardiac and neurological devices, stair-walking wheelchairs, ro- botic surgical systems, and magnetic resonance imaging devices. In contrast

OCR for page 205
0 MEDICAL DEVICES to single-molecule drugs, many complex devices involve a number of com- ponents that, together, form a system.1 Table 7-1 summarizes several ad- ditional differences between devices and drugs as they relate, in particular, to implants and other complex medical devices. In addition to the cost-related differences noted in the table, companies that develop medical devices also have to consider other costs that may be only minor considerations for most pharmaceutical companies. One cat- egory of such costs involves the support and servicing of complex devices once they are released into the market. Depending on the device, highly skilled company personnel may provide training to physicians, clinical staff, and patients (and their families) on the proper use of the device. Service technicians often must be available promptly in case device-related prob- lems arise. Companies must also consider potential obligations to patients if a decision is made at some point to discontinue the device.2 As is true of its products, the medical device industry is likewise quite variable. Some companies are large and have diverse product lines and sub- stantial resources to devote to product development and interactions with government regulators. Compared to the drug industry, a larger proportion of device firms are small, focused on single products and narrow market segments, and limited in their resources (see Gelijns et al., 2006; Linehan et al., 2007). Entrepreneurs at small start-up companies develop many innovative medical devices, including devices that address needs of small patient popu- lations. Company motivations for taking the start-up path to market vary. In some cases, those involved may see the approach as a focused way to address an unmet need and contribute to society without having to navi- gate the decision-making processes of a large, complex company. In some cases, the projected business opportunity is too small or too risky to be worth attention from an existing company but is still attractive enough to attract venture capitalists or a small group of entrepreneurs. In exchange for partial ownership of the start-up company, angel investors and venture 1 An example is a left ventricular assist device for children that FDA approved in 2004. The device consists of four major subsystems—a pump, an external controller, a clinical data ac- quisition system, and a patient home support system—plus accessories, including batteries, a battery charger, and a kit to protect the device during showering (H030003 [the FDA approval number for the device]). Further, the pump subsystem involves multiple elements, including a housing around three additional components—an inflow tube, an outflow element, and a probe to measure blood flow—and a cable connecting the implanted pump to the external battery and controls. 2 For example, implanted devices usually have a finite service life due to battery exhaustion (if electronic or electromechanical) or simply wear and tear, so patients will need replacements. If no alternative device is available and particularly if the patient depends on the device for survival, then the continued availability of a replacement device is crucial. The total replace- ment heart is an example of such a device.

OCR for page 205
0 RARE DISEASES AND ORPHAN PRODUCTS TABLE 7-1 Complex Medical Devices Tend to Differ from Drugs Complex Medical Device Small-Molecule Compound Physical, engineering-based object (or set of Chemical formulation components) Direct mechanism of action and, usually, Indirect biochemical mechanism of action readily apparent, near-term response via blood, other body fluids, or tissue diffusion Site- or organ-specific therapy Usually systemic treatment Patient responses to therapy generally Patient responses variable (benefits and similar and not dependent on dose response adverse effects) and dose dependent High initial product costs amortized over Costs for product accumulate over the service life course of treatment Application often requires professional Application or use is often simple and expertise (e.g., surgical implantation); patient controlled (e.g., taking a pill) patient use might involve complex instructions Continuing product refinement and short Product (basic molecule) not modified, long product life cycle that may improve product life cycle effectiveness and reduce costs Moderate to high development cost High development cost Few basic patents, many incremental Basic patent, fewer incremental patents or patents and products products SOURCES: Adapted from Linehan et al., 2007; Citron, 2008; see also Feigal et al., 2003. Citron, capitalists often provide the financing needed to bring nascent innovations to the market. In addition to infusions of capital, venture capitalists who have worked with other new companies may provide management expertise and strategic advice to guide the managers of a start-up company. As discussed further below, the processes of device development and refinement also differ in significant ways from the processes that character- ize the development of drugs and biologics (see generally Linehan et al., 2007; Pietzsch, 2009; Zenios et al., 2010). Because medical device com- panies are often engaged in a continuous process of product refinement and innovation, patents and similar protections may be less important as a source of competitive advantage for device companies than they are for drug companies. As discussed in Chapter 3, once a new drug is approved by FDA, a pharmaceutical company will have marketplace exclusivity for a specific formulation for a period of time and may also receive patent-term restoration that extends the remaining patent life of the drug. In contrast,

OCR for page 205
0 MEDICAL DEVICES several device companies may compete simultaneously in the marketplace with devices for the same indication that differ in only limited respects. This might be because the devices are not patented or because manufactur- ers have been able to design around the patents that protect a particular competitor’s devices. Consequently, although FDA-approved devices are eligible for patent-term restoration, patents may not be as useful in pro- tecting devices from competition as they are for pharmaceutical products. Even in instances when patents could provide an element of protection from market competition, the patent holder may elect to license its patents to one or more competitors in exchange for royalties or to cross-license patents in order to acquire access to patents held by a competitor. Nevertheless, medi- cal device companies are aggressive in defending their intellectual property from infringements by competitors (Budd and Liebman, 2009). REGULATION OF MEDICAL DEVICES Basic Framework of Medical Device Regulation Although the Federal Food, Drug, and Cosmetic Act of 1938 men- tioned therapeutic medical devices, devices were a relatively inconsequen- tial component of FDA’s jurisdiction. The statute specified that devices be adequately labeled and provide adequate instructions for use but did not give FDA premarket regulatory authority over devices. In the 1970s, fol- lowing widely publicized problems with the Dalkon Shield (an intrauterine contraceptive device) (Hubacher, 2002), Congress turned to the regulation of medical devices with the Medical Device Amendments of 1976 (P.L. 94- 295). The legislation created the basic framework for device regulation. As defined by statute (21 USC 321(h)), a device is an instrument, apparatus, implement, machine, contrivance, implant, in vitro reagent, or other similar or related article, including a component part, or accessory which is: • recognized in the official National Formulary, or the United States Pharmacopoeia, or any supplement to them, • intended for use in the diagnosis of disease or other conditions, or in the cure, mitigation, treatment, or prevention of disease, in man or other animals, or • intended to affect the structure or any function of the body of man or other animals, and which does not achieve any of its primary intended purposes through chemical action within or on the body of man or other animals and which is not dependent upon being metabolized for the achievement of any of its primary intended purposes.

OCR for page 205
0 RARE DISEASES AND ORPHAN PRODUCTS Within FDA, the Center for Devices and Radiological Health (CDRH) regulates most medical devices. The Center for Biologics Evaluation and Research regulates devices related to blood and cellular products such as blood collection, screening, and processing devices. The OOPD has roles in designation of devices eligible for HDE approval and in awarding product development grants, which are available for device as well as drug development. Device Classification and Regulation A fundamental element of the 1976 law was a risk-related device clas- sification scheme that forms the basis for risk-related regulatory require- ments. To simplify, the law designated devices of lowest risk and relatively little complexity as Class I; devices of moderate risk and greater complexity as Class II; and devices that support or sustain life or otherwise present a high risk to the patient as Class III.3 In general, Class I and II devices have substantially equivalent prede- cessor or “predicate” devices that are already on the market. Some new devices may be classified automatically as Class III devices because they have no such predicate device. FDA may reclassify such devices as Class II devices based on an analysis of the risk they present. For example, such a reclassification was requested for the first device available to screen new- born infants for inherited abnormalities of amino acids and deficiencies in certain enzymes (Lloyd, 2004).4 Regardless of its complexity, any device can present potential harms to patients if it is misused, mislabeled or poorly labeled, badly designed, poorly manufactured, or misrepresented. Thus, the regulatory framework created by Congress covers all classes of devices and extends, in some cases, to requirements for sponsors to conduct postmarket studies to collect data about safety and effectiveness after a device is approved for marketing. For Class I devices, manufacturers generally must register with FDA and follow FDA’s quality system regulations, including adherence to good manufacturing practices. These devices are usually not subject to premarket notification or review. Manufacturers of Class II devices usually must get FDA clearance of a “510(k) notification” (named for the relevant section of the law) to legally market these devices. The process requires the submission of considerable technical information and sometimes animal study data related to safety and performance characteristics of the new device, in order to demonstrate its “substantial equivalence” to the predicate device. The 510(k) notifica- 3 As of 2006, Class I, II, and III devices accounted for approximately 43, 45, and 13 percent of classified devices, respectively (Tillman as cited in IOM, 2006, p. 76). 4 This was the NeoGram Amino Acids and Acylcarnitine Tandem Mass Spectrometry Kit.

OCR for page 205
 MEDICAL DEVICES tion typically does not require clinical data unless the technology for a new device differs from that of the predicate device and clinical data are neces- sary to evaluate the potential impact of this difference on safe and effective performance. Clinical data are included in approximately 10 percent of 510(k) notifications (Tillman and Gardner, 2004; Rosecrans, 2010). Both a CDRH working group and an Institute of Medicine (IOM) committee (as requested by FDA) are evaluating aspects of the 510(k) process. The CDRH group issued its preliminary report in August 2010 (CDRH, 2010a). For Class III devices, which account for a small proportion of all legally marketed medical devices, manufacturers must submit premarket approval (PMA) applications and provide data from clinical trials to demonstrate reasonable assurance that a device is safe and effective (what this report terms efficacy) for the intended use in the intended patient population. Ex- amples of Class III devices include implanted devices such as the titanium rib, some diagnostic test kits, and certain surgical sealants. Securing FDA approval of such a device is usually complex, costly, and time-consuming, taking on the order of several years. The cost will vary depending on the complexity of the device and the kinds of nonclinical and clinical data that the sponsor must submit to demonstrate safety and efficacy. This report focuses on complex devices intended specifically to treat complex rare conditions. Most will be Class III devices and thus will require formal authorization by FDA, usually through the PMA process or, in some cases, the 510(k) process. For qualifying devices intended for a small population, approval can also come through the HDE process described below. The committee is not aware of any analysis that attempts to catalog devices that have been cleared under the 510(k) process or approved under a PMA specifically for the treatment of rare conditions defined according to the Orphan Drug Act (i.e., conditions affecting fewer than 200,000 individuals). It has found examples of such devices. For example, for the rare eye condition kerato- conus, CDRH has cleared devices under the 510(k) process (K992466 and K024164) and also approved a different type of device for the condition un- der an HDE (H040002). At least one HDE-approved device, Bioglue, was subsequently approved for broader indications under a PMA (P010003). For devices that are designated as “significant risk devices” because they have the potential to cause serious harm to research participants, man- ufacturers must secure an Investigational Device Exemption (IDE) before they can conduct clinical studies in humans with the devices.5 Similar to the Investigational New Drug application, an IDE application must include 5 “Nonsignificant-risk” device clinical trials require institutional review board approval and informed consent but not an approved IDE application (21CFR 812.2(a), 812.3(m)). They are subject to the “abbreviated IDE” requirements and are considered to have a “deemed ap- proved” investigational device exemption (21 CFR 812.2(b)(1)).

OCR for page 205
 RARE DISEASES AND ORPHAN PRODUCTS data about preclinical studies and any available clinical information. It must also provide a description of the proposed research and analysis strategy. An IDE may prompt extensive discussions and negotiations between the manufacturer and FDA to arrive at agreement on a research plan that will provide data of acceptable quality to support FDA approval of the device. As described in one review of the process, the “first and arguably most im- portant step in this process is the pre-IDE meeting, in which the company, often accompanied by the lead clinical investigator(s), meets with FDA/ CDRH to present data about the device, its clinical development program, and its intended use after approval” (Kaplan et al., 2004, p. 3071). As is the case for pharmaceuticals, FDA may approve medical devices with requirements for postmarketing studies, including clinical studies. For example, when CDRH approved a transcatheter pulmonary valve system under an HDE, it required two postapproval clinical studies (Tillman, 2010). CDRH now tracks the status of postapproval studies required after January 1, 2005, and posts tracking information on a public web page. Diagnostic Devices, In Vitro Devices, and Genetic Tests FDA regulates a range of diagnostic devices under the procedures de- scribed above. Diagnostic devices include such diverse items as blood pres- sure cuffs, vision evaluation instruments, cardiac monitors, and sophisticated imaging equipment. Based on their complexity, diagnostic devices are gener- ally assigned to one of the three classes discussed above and regulated ac- cordingly. CDRH has approved HDEs for three diagnostic testing devices.6 Diagnostic devices also include an array of products known as in vitro diagnostic devices, which “are those reagents, instruments, and systems in- tended for use in diagnosis of disease or other conditions” (21 CFR 809.3(a)). FDA regulates in vitro diagnostic devices that are developed and sold by device manufacturers as test kits. In vitro diagnostic devices include genetic and other tests that are important in diagnosing many rare diseases.7 6 These are the Fujirebio Mesomark Assay (H060004, approved in 2007), the Heartsbreath test (H030004, approved in 2004), and the TAS Ecarin Clotting Time Test (H990012, ap- proved in 2000). 7 FDA recognizes two categories of in vitro devices that are in research stages. “Research use only” products are devices that are used only in the preclinical “laboratory research phase of development, that is, either basic research or the initial search for potential clinical utility” (CDRH-CBER, 2007, p. 12). Such a device may not be used for human clinical diagnostic or prognostic use, and the labeling must state: “For research use only. Not for use in diagnostic procedures” (21 CFR 809.10(c)(2)(i)). “Investigational use only” devices are products that are in “the clinical investigation stage of development” but that may be exempt from IDE require- ments (CDRH-CBER, 2007, pp. 12-13. These products must be labeled: “For Investigational Use Only. The performance characteristics of this product have not been established” (21 CFR 809.10(c)(2)(ii)). See also Gibbs (2010).

OCR for page 205
 MEDICAL DEVICES In addition to using in vitro diagnostic test kits to perform diagnostic testing, some clinical laboratories develop their own in-house assays, known as laboratory-developed tests. These laboratory-developed tests are cur- rently regulated under the Clinical Laboratory Improvement Amendments of 1988 (CLIA) and state laws (Maloney, 2010). With rare exceptions, laboratory-developed tests usually are not regulated by FDA.8 Recently, however, FDA announced its intention to regulate all laboratory-developed tests as medical devices, as discussed below. FDA regulations do, however, require that a clinical laboratory that develops a test using an analyte-specific reagent must disclose its regulatory status and must add a statement on test reports that the test has not been cleared or approved by FDA (21 CFR 809.30(e)). Analyte-specific reagents (which include polyclonal and monoclonal antibodies, specific receptor pro- teins, nucleic acid sequences, and similar reagents) are the building blocks that clinical laboratories use to develop in-house assays. Also, although laboratory-developed tests themselves are not usually regulated by FDA, analyte-specific reagents are regulated as “restricted devices.” Manufac- turers cannot make any claim of clinical or diagnostic effectiveness for an analyte-specific reagent and can only describe what substance it will identify. If a manufacturer combines analyte-specific reagents into a kit, or otherwise offers them for sale together, then the product must be approved as a medical device. Because most genetic tests are available only as laboratory-developed tests, they are not regulated by FDA (Huang and Javitt, 2008). In a report on the regulation of genetic tests prepared by the Secretary’s Advisory Com- mittee on Genetics, Health, and Society, the group identified shortcomings in several areas, including regulations governing clinical laboratory quality and “oversight of the clinical validity of genetic tests” (SACGHS, 2008, p. 191). In the past several years, various groups have recommended that FDA should regulate either all genetic tests or all laboratory-developed tests under the medical device authorities (Mansfield and Tezak, 2010). In June 2010, CDRH announced a public meeting and requested comments on is- sues related to the regulation of laboratory-developed tests (75 Fed. Reg. 34463). Although the agency has indicated that it plans to regulate some of these tests as medical devices, the specifics and priorities have yet to be decided. In noting the challenges of encouraging innovation while assuring 8 In draft guidance, CDRH identified one category of laboratory-developed tests, the in vitro diagnostic mulitvariate index assay, as subject to its regulation (CDRH-CBER, 2007). It stated that such “tests are developed based on observed correlations between multivariate data and clinical outcome, such that the clinical validity of the claims is not transparent to patients, laboratorians, and clinicians who order these tests” and “frequently have a high risk intended use” (p. 4). Included in this category are tests that integrate genetic and other information to predict a person’s risk of developing a disease.

OCR for page 205
 RARE DISEASES AND ORPHAN PRODUCTS the safety and efficacy of laboratory-developed tests, the CDRH announce- ment specifically cited tests for rare conditions. Another area of regulatory complexity is co-development of a drug and a companion diagnostic. An example is a diagnostic test kit to assess whether a breast cancer patient has a gene mutation that is targeted by the drug trastuzumab (Herceptin). FDA held up approval of the drug until an approved in vitro diagnostic could be substituted for the laboratory- developed test that was initially used in clinical trials. It approved both the drug and the diagnostic in 1998 (Madsen, 2004). After a 2005 concept pa- per on the topic generated considerable criticism (see, e.g., PMC, 2009), the FDA Commissioner indicated that a new draft guidance document would be published in 2010 and would reflect public comments and scientific and other developments (Hamburg, 2009; Ray, 2010). In the meantime, FDA has been applying a case-by-case approach to regulation of companion diagnostics (Carver, 2010). Combination Products Some medical products combine a medical device and a drug or biologic. Examples include the drug-eluting coronary stent (which adds a drug coating to a metal stent in order to reduce the risk of reocclusion of the coronary artery) and the fentanyl patch (which delivers the drug through the skin). Combinations can take several different forms. For a product such as the drug-eluting stent, the device and drug components are truly combined into a single entity. Two items that are physically distinct but packaged together also qualify as a combination product. The category can also cover a product such as a drug that is packaged separately but is labeled as being for use only with a specific device or type of device (such as a specific diagnostic test).9 At least one combination product has been approved by CDRH through the HDE process (OP-1 Putty under H020008).10 As discussed below, the 9 The Office of Combination Products, which was created in 2002, assigns primary respon- sibility for regulating combination products to the most appropriate unit of FDA. As a general rule, that assignment is based on the primary mode of action of a combination product. Thus, a drug-eluting stent is intended primarily to open a blood vessel, with the drug activity second- ary, so regulation would be assigned to CDRH. In contrast, the Center for Drug Evaluation and Research (CDER) would take the regulatory lead for a drug-eluting disk that is intended to deliver chemotherapy agents for brain tumors. For some combinations, the lead might go to the Center for Biologics Evaluation and Research. 10 The product is approved under the HDE only for use in the posterolateral (intertrans- verse) lumbar spine in a limited patient population. It is made from mixture of a genetically engineered human protein powder, bovine collagen, saline solution, and a thickening agent to form a putty-like material that is applied to each side of the spine section that is to be fused. In 2008, FDA notified health care practitioners of reports of life-threatening complications as- sociated with one of these elements, recombinant human bone morphogenetic protein (rhBMP)

OCR for page 205
 MEDICAL DEVICES different incentives for the development of orphan drugs and for the de- velopment of devices for small populations theoretically could complicate collaboration on combination products for small populations. Alternate Approval Route for Medical Devices for Small Populations As is true for companies that manufacture drugs and biologics, device companies naturally seek business opportunities in markets of sufficient size and profitability to warrant the investment risk. Particularly if FDA requires extensive clinical data for approval of the device, companies may be discouraged from pursuing devices for small markets by the expense and practical challenges of conducting acceptable trials to demonstrate safety and effectiveness. The Safe Medical Devices Act of 1990 authorized the Humanitarian Device Exemption to encourage the development and introduction of com- plex device technologies to meet the needs of small patient populations. Although neither the text nor the title of the 1990 law uses the term “rare disease” or “orphan product,” the purpose is broadly similar to the pur- pose of the Orphan Drug Act. The specifics vary in part because the details of device regulation differ and in part because the incentives (particularly market exclusivity) that were viewed as important for drug manufacturers were viewed as less meaningful for device manufacturers. An HDE application is the same as a PMA application except that it need not include evidence of effectiveness, a characteristic that also distin- guishes the requirements for an HDE from the requirements for FDA ap- proval of an orphan drug. The HDE application must, however, “contain sufficient information for FDA to determine that the device does not pose an unreasonable or significant risk of illness or injury, and that the prob- able benefit to health outweighs the risk of injury or illness from its use” (CDRH, 2009, unpaged). To be eligible for an HDE, a manufacturer must first request that the device be designated by the OOPD as a Humanitarian Use Device (HUD). A HUD is a “medical device intended to benefit patients in the treatment or diagnosis of a disease or condition that affects or is manifested in fewer than 4,000 individuals in the United States per year” (21 CFR 814.102(a)(5)). (If a device is for diagnostic purposes, the documentation in an HDE ap- plication must demonstrate that fewer than 4,000 patients per year would be subjected to diagnosis by the device in the United States.) The statutory language has caused some confusion about whether it refers to incidence or when used in the cervical spine (Schultz, 2008). The manufacturer was indicted in 2009 on charges of illegal promotion of the product for unapproved uses (DOJ, 2009).

OCR for page 205
0 RARE DISEASES AND ORPHAN PRODUCTS disorders. Deep-brain stimulators were eventually approved by FDA for several new indications, each involving relatively minor technical changes to the device. The implant received PMA approval for essential tremor in 1997 and advanced Parkinson disease in 2002 and then was granted HDE approval for dystonia in 2003 and chronic, severe, treatment-resistant ob- sessive compulsive disorder in 2009. University researchers have been actively involved in developing the technologies used in the genetic and drug discovery research described in earlier chapters. The American Institute for Medical and Biological Engi- neering cites the development of genomic sequencing and microarrays in its “hall of fame” (AIMBE, 2005). The bioengineering program at Stanford University, summarized in Box 7-3, offers one example of the intersection of device engineering and scientific advances in biological sciences. Although some technological advances in medical devices have taken advantage of government-supported research and development, direct gov- ernment investment in accelerating medical device research and devel- opment was initially limited. One exception is the total artificial heart BOX 7-3 Stanford University Biodesign Program About 10 years ago, faculty at Stanford University developed a systematic ap- proach to solving significant medical problems in which invention and innovation were a team activity and were part of a process. The impetus for this initiative, the Stanford Biodesign Program, was the realization that innovations in the medical area involved many technical and scientific disciplines and these disciplines need to collaborate and inform each other. Biodesign is associated with Stanford’s Bio-X program, which promotes in- terdisciplinary research in biology and medicine. As described on that program’s website, “Ideas and methods embodied in engineering, computer science, phys- ics, chemistry, and other fields are being brought to bear upon important chal- lenges in bioscience. In turn, bioscience creates new opportunities in other fields. Significant discoveries and creative inventions are accelerated through formation of new collaborative teams” (http://biox.stanford.edu/about/index.html). The Biodesign Program creates multidisciplinary collaborative teams com- posed of graduate students from engineering, medicine, and business. These teams follow a three-stage process or method to create cost-effective, state- of-the-art medical devices for the benefit of patients, industry, and society. The method includes three stages: need identification; concept development; and business or project planning. The program has been emulated both domestically and internationally. Stanford faculty have published a textbook Biodesign (Zenios et al., 2010) that describes the process.

OCR for page 205
 MEDICAL DEVICES program at the National Heart, Lung, and Blood Institute (NHLBI), which began in 1964 (IOM, 1991). Although the program’s focus shifted to ven- tricular assist devices (which have a larger target population than the arti- ficial heart per se), the initial investments provided important knowledge for the development of less ambitious but clinically relevant cardiac support systems. The report of an expert panel convened by NHLBI noted that it “is not probable that development of assist devices would have occurred without the government support that is now being increasingly assumed by industry as clinically effective devices move towards marketing approval” (NHLBI, 2000, p. 3). During the 1970s, the engineering community sought to increase the visibility of biomedical engineering and to educate the National Institutes of Health (NIH) about the nature and value of research in bioengineering and bioimaging (Hendee et al., 2002). In 2000, after a number of unsuc- cessful proposals in the 1990s, Congress created the National Institute of Biomedical Imaging and Bioengineering (NIBIB, 2009). NIBIB seeks to advance basic research and improve patient care by integrating the physical and engineering sciences with the life sciences. Relevant disciplines extend beyond biological sciences and various engineering disciplines to include (among others) the information sciences, physics, chemistry, mathematics, materials science, and computer science. The range of NIBIB interests cov- ers, among other areas, biomaterials, bioinformatics, structural biology, drug and gene delivery systems and devices, tissue engineering, microbio- mechanics, nanotechnology, sensors, surgical instruments, diverse kinds of imaging, and rehabilitation technology. Some areas of device innovations that illustrate the interaction between innovations in engineering and bio- logical sciences are summarized in Box 7-4. In addition, the discussion in Chapter 4 of discovery research and diagnostic developments identifies other areas in which scientific and technological advances in biomedicine will likely shape innovation in diagnostic devices. Clinical Studies for Medical Devices As discussed earlier, FDA clears the majority of medical devices for marketing without requiring formal clinical studies. For a small proportion of devices—in particular, implanted devices or other high-risk devices— FDA requires data from clinical studies. For significant-risk devices, FDA requires formal approval of a request to begin clinical studies. Under its regulations governing clinical studies to support PMAs, FDA states that the agency relies only on “valid scientific evidence to determine whether there is reasonable assurance that the device is safe and effective” (21 CFR 860.7(c)). The same document goes on to describe valid scientific evidence as

OCR for page 205
 RARE DISEASES AND ORPHAN PRODUCTS BOX 7-4 Innovations in Engineering and Biological Sciences and Medical Device Innovation Replacement organs: Using man-made scaffolds and other biomedical tech- niques, tissue engineering and regenerative medicine discoveries will permit scientists to “grow” organs in the laboratory to replace patients’ failed organs. Tissue-engineered urinary bladders have already been implanted in patients. Proof-of-principle laboratory demonstration of a total beating heart (rat) has also been achieved. Such technologies have the potential to overcome the shortage of available donor transplants and to offer patients a biological tissue solution rather than an electromechanical therapy. Drug delivery: Oral delivery of drugs has significant limitations, including patient compliance, first-pass inactivation by the liver, systemic rather than site-specific effects, and undesirable variability of blood levels between doses. Early genera- tions of implanted drug delivery systems for chronic diseases or symptoms such as spasticity and intractable pain have demonstrated capabilities that address many of the limitations cited for oral delivery. Future generations are expected to provide sensor-based closed-loop operation, delivering drugs only when needed and at appropriate dosages. Delivery will be either site-specific, treating only affected tissue, or systemic. Of particular interest is the development of a fully implanted artificial pancreas that will deliver appropriate amounts of insulin in response to continuous monitoring of diabetic patient’s blood glucose levels. Implanted diagnostics: The concept of an implanted laboratory-on-a-chip has the possibility of revolutionizing how disease is diagnosed and preventing certain diseases or crisis episodes. Biosensors, as described for implanted insulin pumps above, serve as an example of how this might work. Implanted “chips” that contain an array of sensors will continually monitor a patient’s condition and, if warranted, communicate to the patient’s health care provider that attention is needed. Pro- totypic versions of rudimentary diagnostic systems have already demonstrated the ability to reduce serious events related to heart failure and also to reduce the number of visits to the emergency room. Miniaturization: A collection of enabling technologies, some emerging from the field of nanotechnology, will expand possibilities of minimally invasive surgery, pro- duce novel bio-interactive coatings, and reduce the size and expand the service life of implants, making them more suitable for pediatric patients. SOURCES: AdvaMed, 2004; Braunschweig, 2007; El-Khatib et al., 2010; Trafton, 2010. evidence from well-controlled investigations, partially controlled studies, studies and objective trials without matched controls, well-documented case histories conducted by qualified experts, and reports of significant human experience with a marketed device, from which it can fairly and responsibly be concluded by qualified experts that there is reasonable as-

OCR for page 205
 MEDICAL DEVICES surance of the safety and effectiveness of a device under its conditions of use. The evidence required may vary according to the characteristics of the device, its conditions of use, the existence and adequacy of warnings and other restrictions, and the extent of experience with its use. Isolated case reports, random experience, reports lacking sufficient details to permit sci- entific evaluation, and unsubstantiated opinions are not regarded as valid scientific evidence to show safety or effectiveness. The requirements for evidence to support FDA approval of a PMA explic- itly provide for more variability (linked to the particular characteristics and uses of a device) than is found in the corresponding expectations outlined in Chapter 3 for the approval of drugs. A recent study examined FDA sum- maries of the evidence used to support FDA approvals for 78 “high-risk cardiovascular devices” (Dhruva et al., 2009). The analysts reported that nearly two-thirds (65 percent) of the applications were approved based on a single study, that 27 percent of the 123 submitted studies were randomized, that 88 percent of the primary endpoints used were surrogate measures, that 52 percent of the endpoints were compared with controls, and that 31 percent of these controls were historical. Moreover, as described earlier, FDA does not require the same level of evidence for a device approved through an HDE as it does for those approved through a PMA. For medical devices that require clinical evaluation, a pilot or feasibil- ity study usually involves an initial clinical evaluation of the safety of a prototype device in individuals with the condition for which the device is designed. Such a study may suggest modifications to the prototype device to improve its performance. For a device that requires complex surgery for its implantation or that otherwise is technically demanding to use, the pilot phase may also provide an important period for learning about the process and skills required for the safe and effective clinical use of the device. Ex- perience gained from pilot studies also contributes to the design of pivotal studies, which usually recruit larger numbers of research participants and may involve multiple study sites and centers. If surgical procedures are involved, the process may also require training of investigators at sites not involved in the pilot study. As is also the case for orphan drugs, the accumulation of sufficient par- ticipants may take years for a device that is intended for a small population of patients. For example, clinical testing for the titanium rib (cited at the start of this chapter) occurred over 14 years, a long period that reflected in part the challenges of working with a rare condition and in part the request by FDA for long-term information on the device, which requires repeated adjustment as a child grows (Campbell, 2004). In addition, many devices present special challenges for the design of clinical trials. Especially for surgically implanted devices, the classic ran-

OCR for page 205
 RARE DISEASES AND ORPHAN PRODUCTS domized, double-blind comparative study is often not feasible or ethical. A few trials of surgical implant procedures have included sham surgeries for comparison in single-blind studies, but the surgical team obviously had to be aware of which procedure was used (see, e.g., Moseley et al., 2002). Such trials are controversial (see, e.g., Miller, 2003; Mehta et al., 2007). For some electronic devices such as cardiac or neurological stimulators, clinical studies have sometimes used a design that involves implantation of the device in a study population and then comparing a subset of the group in which the device is kept switched on with another subset in which it is switched off for a predetermined period of time (see, e.g., Greenberg et al., 2006). Staff at CDRH are planning two projects that should provide a better understanding of issues in the design of clinical studies for devices (Linda C. Ulrich, M.D., Medical Officer, FDA Office of Orphan Product Develop- ment, April 26, 2010, personal communication). One project is an analysis of the clinical safety and efficacy data submitted in support of PMAs. The other project is preparation of a guidance document on clinical trial design for device trials. In addition, reflecting the characteristics of device trials, CDRH has developed guidance on the use of Bayesian statistics with clinical trials of medical devices, including situations involving confirmatory trials, device modifications, incomplete data, and opportunities for adaptive design strat- egies. (FDA released draft guidance in 2006 and final guidance in 2010 [CDRH-OSB, 2010].) FDA development and education efforts in this area date back well over a decade (Campbell, 2008, 2009). Although the Bayes- ian approach requires companies to have expert statistical advice and to engage in early consultation with FDA, it has the potential to reduce the costs of trials. As of 2009, at least 20 PMAs or PMA supplements using Bayesian analysis were under review (Campbell, 2009). As part of the guid- ance and education programs on small clinical trials that NIH and FDA are conducting, a commentary examining applications of Bayesian statistics to device trials involving small populations could be useful. In general, recent years have seen a growing appreciation of the special challenges of device trials, the importance of innovative trial and statistical methods, and the opportunities for FDA-industry interaction to improve trial design and analysis techniques and their use (Campbell, 2008). DEVICE INNOVATION AND THE HDE OPTION Although individuals and companies may pioneer devices for rare dis- eases for purely altruistic reasons, such altruism is uncommon because device innovation entails significant costs, protracted time for research and development, and commitment to ongoing support and administrative costs once a device is approved. For complex “new to the world” devices, the

OCR for page 205
 MEDICAL DEVICES research and development costs can run into the tens of millions of dollars, although details of these costs are not readily available. The time lines to produce practical, safe, and reliable implanted devices can be very long, measured in years and even decades. An HDE approval reduces the time to market because demonstration of effectiveness is not required. Companies can also recover certain costs, for example, research and development costs. Nevertheless, without a reason- able opportunity to make a profit, as provided by the Orphan Drug Act, the costs and investment risks to bring a new technology forward for small markets are substantial and are likely not to appeal to many companies and investors. These reservations may be moderated if development of a device for a small population is considered a stepping stone for a future application that may serve a larger market. In situations where the technology is truly novel, the HDE can offer a company the opportunity to learn more about it while continuing its development for a broader use. For example, the company Spiration, Inc., received HDE approval for a device to control prolonged air leaks in the lung following lung surgery (H060002). The company is also conducting clinical testing of the device to treat severe emphysema. In anticipation of additional indications for a broader patient popula- tion, a company may view an HDE approval as a way to enter the market more quickly and with a baseline level of revenue. Market entry under an HDE provides a company with an opportunity to further evaluate the technology and identify next-generation design improvements. It also gives surgeons an opportunity to refine surgical techniques and protocols that may benefit future patients. In addition, an HDE could have “good will” value to the sponsor. Nevertheless, as noted above, there remain both the direct costs of supporting the HDE device and the allocation of financial and personnel resources (especially difficult for a small company) to the HDE device in lieu of another IDE device, another indication, or other research and development effort. Some reservations about the HDE option may be moderated if a com- pany is looking at a new indication for an already approved device. The list of HDE devices includes some devices (e.g., the deep-brain stimulation device) that are modifications of existing products approved under PMAs for other indications. The investment risk to the company for pursuing a new rare diseases indication is moderated in such cases because most of the research and development costs intrinsic to the device have already been incurred. Companies will still incur some additional incremental research and development investments to devise any modifications needed for the rare diseases application and to generate data on safety and probable ben- efit necessary for the HDE application. Although the incremental costs may be relatively modest for an HDE

OCR for page 205
 RARE DISEASES AND ORPHAN PRODUCTS based on an already approved product, companies still face the limit on profits for HDE devices and uncertainties about reimbursement. They also still must consider the opportunity cost for pursuing an HDE approval rather than pursuing development of other products or pursuing approval through the regular PMA process. A company’s decision about the HDE op- tion could also be influenced by the requirements for IRB approval and the potential for the annual market for the product to be larger than projected and exceed the annual shipment limit. In the latter situation, FDA would likely ask the company to withdraw the HDE device and seek approval through the PMA process. Interviews conducted by Bernad (2009) suggest that device develop- ers may sometimes decide to pursue HDE approval after the major part of product development has occurred. One company representative noted that the HDE process can save 3 to 4 years in getting a product to market, “which can be the entire product life cycle” (p. 140). An executive for a company with a device that has both PMA and HDE approvals observed that the HDE process was a means of broadening approved indications for the device that “saved 3 years, recovered $10 million from the initial research and development costs, and established good relationships with many physicians in the field” (p. 142). Others interviewed noted that the short-term benefits of the HDE process must be weighed against the nega- tives of the restriction on profits and the potential for insurers not to cover the device (for lack of evidence of efficacy). One official particularly cited the burdensome IRB process as involving substantial costs for meticulous record keeping, application production, and IRB fees (which could involve hundreds of sites). In addition, in an article summarizing the results of a symposium on the HDE process, Kaplan and colleagues noted that the availability of a device through the HDE process could complicate recruit- ment for clinical studies to evaluate the safety and efficacy of a device for a more common indication (Kaplan et al., 2005). As noted earlier, the difference between drug and device incentives could be an issue for a combination product for a small population if the difference in incentives discouraged drug-device company cooperation on a combination product that involved a complex device and that did not intrinsically require simultaneous or coordinated clinical testing of the drug and the device. If such situations arise, Congress or FDA may need to consider a modification in the HDE policy to encourage innovation to meet unmet needs for combination devices while protecting patients from unsafe or ineffective products. In general, however, rare conditions may be treatment targets for either a drug solution or a device solution but not both. That is, only one or the other modality holds clinical promise or clinical relevance or presents a reasonable risk-benefit ratio. In such instances, inconsistencies in incen-

OCR for page 205
 MEDICAL DEVICES tives and regulatory requirements for orphan drugs and Humanitarian Use Devices are not likely to have much practical impact on the development or use of the product. In addition, if a device is one that qualifies for clearance through the 510(k) process, then the HDE process and conditions likely will not be considered.20 It is difficult to assess the extent to which the relatively low number of HDE approvals is influenced by the profit disincentive and the limit on yearly shipment or is, rather, a function of limited opportunities for devices—compared to drugs—to meet substantial unmet health needs for small populations. As noted at the beginning of this chapter, the emphasis in discussions of rare diseases is overwhelmingly on drugs. For devices covered by an HDE, information on the number of device units shipped is not readily available nor are the estimates submitted by companies (in support of their HDE application) of the number of af- fected individuals. Although indication-specific information is not available, a recent press release by Medtronic recently reported that a cumulative 75,000 patients worldwide had been treated with its implanted deep-brain stimulation technology, including for the four indications described earlier (Medtronic, 2010). RECOMMENDATIONS Based on information presented to the committee by representatives of companies and FDA and provided by a review of past proposals for policy change, the committee concluded that the development of reasonable and effective incentives specific to device development for small populations has proven difficult. The incentives relevant for drug development, par- ticularly the protections from market competition, are not well matched to the realities of device development. Although recent initiatives to promote the development of pediatric medical devices modify the HDE process by allowing profits, they do not move toward either the marketing protections or the stricter approval requirements applicable to orphan drugs. In contrast to pediatric medical devices, relatively little attention has been directed to needs for medical devices for people with rare conditions. (Even the statement to the committee from the Advanced Medical Technol- ogy Association focused as much or more on pediatric devices as on devices 20 The committee found one rare condition for which FDA has approved an orphan drug and an HDE, although the approved indication for the device appears considerably more restrictive. In 2000, the agency approved botulinum toxin type A as an orphan drug to de- crease the severity of abnormal head position and neck pain associated with cervical dystonia (BLA 103000/1004). In 2003, it approved an implanted deep-brain stimulation device under an HDE for the management of chronic, severe, drug-refractory dystonia, including cervical dystonia (H020007).

OCR for page 205
 RARE DISEASES AND ORPHAN PRODUCTS for rare diseases [AdvaMed, 2009].) CDRH recently held a meeting to discuss unmet devices needs, but it did not specifically address the needs of people with rare conditions. Without the time for a very focused examination, the committee found it difficult to assess the possible extent of unmet device needs for adults with rare conditions and the extent to which changes in FDA policies (e.g., an increase in the criterion of 4,000 patients per year for HDE approval) might promote innovation to meet these needs). A first step in understanding the potential areas for device innovation is a needs assessment for adults with rare conditions. Such an assessment, which should involve patient groups, clinicians, biomedical engineers, and device developers, can also illuminate impediments to innovations to meet those needs. RECOMMENDATION 7-1: FDA and NIH should collaborate on an assessment of unmet device needs and priorities relevant to rare diseases. That assessment should focus on the most plausible areas of unmet need, identify impediments to meeting these needs, and exam- ine options for overcoming impediments and stimulating high priority innovations. The identification of needs, priorities, and impediments should help inform the consideration by government, private foundations, and others of additional incentives and supports for medical device development for small populations. Beyond simplifying some aspects of the HDE process as suggested in Recommendation 7-4, options to encourage device innovation for rare diseases include • the provision of additional orphan products grants and NIH awards for the development of devices to meet priority needs; • the authorization of tax credits for certain research and develop- ment costs, similar to those available for companies developing orphan drugs; and • the creation of inducement prizes for the design and initial testing of novel devices in areas of unmet need. In addition, the changes in the HDE incentives for pediatric devices, including the relaxation of the restriction on profits, provide an opportunity to examine the case for similar changes to encourage innovative devices for adults with rare conditions. Also, as experience with the revised incentives for pediatric devices is gained, including through the GAO evaluation due in 2010, that knowledge should be applied to the encouragement of devices for adults with rare conditions. One alternative to eliminating the profit restriction altogether would be the development of a cost-plus option that

OCR for page 205
 MEDICAL DEVICES would allow companies to charge a specified amount over certain costs of development. The committee did not examine this idea in depth and recognizes that it would need careful investigation of potential unintended consequences and consideration of safeguards. RECOMMENDATION 7-2: Congress should consider whether the rationale for its creation of additional incentives for pediatric device development also supports the use of such incentives to promote the de- velopment of devices to meet the needs of adults with rare conditions. A modest step to encourage additional company interest in devices for small populations would involve greater flexibility in the limits on annual shipments of HDE devices. For devices covered by an HDE, information on the number of units shipped is not readily available nor are the esti- mates submitted by companies (in support of their HDE application) of the number of affected individuals. Such information might help in assessing how often the 4,000-per-year shipment limit is approached and thus how often the limit might restrict access within the existing framework of HDE policy. Such information would not, however, help in determining what applications might be attracted by a higher cap given that the HDE unlike the PMA does not require demonstration of efficacy. Rather than authoriz- ing an increase in the yearly limit of device shipments that are allowed for an HDE device, Congress could provide for the cap to be raised in specific situations when CDRH determined (based on medical, demographic, and scientific information submitted to it) that an increase in the cap or an- nual distribution number would benefit patients with a rare disease. This policy would require the specification of boundary criteria (e.g., how large a deviation in the cap could be approved). An analysis of experience with company estimates of affected populations and annual shipments could help in evaluating this idea. RECOMMENDATION 7-3: As a basis for possible congressional ac- tion, the Center for Devices and Radiological Health should analyze the supporting justifications offered in successful and unsuccessful Human- itarian Device Exemption applications related to the 4,000-person-per- year limit and should evaluate the subsequent experience with actual device shipments for approved applications, including any communica- tions about projections that a company might exceed the limits. Taking the findings into account, Congress should consider authorizing FDA to permit a small, defined deviation from the yearly limit on shipments for a specific device when the agency determines that such a deviation would benefit patients with a rare disease.

OCR for page 205
0 RARE DISEASES AND ORPHAN PRODUCTS In addition, FDA could take other steps to make the HDE process worth further consideration by potential sponsors. It could offer additional guidance and assistance to make it easier for sponsors and IRBs to manage the requirement for IRB review of HDEs. As discussed earlier, recent CDRH guidance is helpful but still leaves areas of confusion and uncertainty. The agency could also clarify existing guidance on the very specific details of the process for applying for designation of a Humanitarian Use Device. For example, it could provide further guidance on the evidence needed to sup- port claims of probable benefit and the calculation of the size of the target patient population, and it could also offer consultation to help sponsors understand what data will be responsive and justifiable. RECOMMENDATION 7-4: FDA should take steps to reduce the bur- dens on potential sponsors of Humanitarian Use Devices, including • assigning an ombudsman to help sponsors navigate the regula- tory process for these applications; • providing more specific guidance and technical assistance on the documentation of the size of the patient population as required for humanitarian use designations; and • developing better guidance (including step-by-step instructions and sample documents) for sponsors and IRBs on their roles and re- sponsibilities related to IRB review of HDEs. In another area, CDRH could also develop new guidance on the use of surrogate endpoints in medical device trials (see discussion in Chapter 5). As noted earlier in this chapter, the planned analysis of the clinical safety and efficacy data submitted in support of PMAs and the planned guidance document on device clinical trials could contribute useful information and perspectives in this area, even though neither activity will focus on HDEs specifically.