International Journal of Technology Assessment in Health Care, 8:4 (1992), 583-597. Copyright © 1992 Cambridge University Press. Printed in the U.S.A.

CREATING THE COSTLIEST ORPHAN The Orphan Drug Act in the Development of Ceredase™ Dana P. Goldman Stanford University

Ann E. Clarke Stanford University

Alan M. Garber Palo Alto Department of Veterans Affairs Medical Center and Stanford University

Abstract The FDA recently approved Ceredase", a new treatment for Gaucher's disease, under the provisions of the Orphan Drug Act. Ceredase™ is unusually expensive, but there are no satisfactory alternative therapies. It appears likely that Ceredase™ would not have become available without the protection of the Orphan Drug Act, but its expense and the lack of information about its long-term effects on health raise questions about whether the ODA provides appropriate incentives to develop cost-effective technologies.

The drug development and approval process is time-consuming and potentially costly enough to preclude the development and marketing of treatments for rare disorders. The Orphan Drug Act (ODA) contains provisions that are designed to overcome critical regulatory barriers to the development of such treatments. These provisions include research grants, investment tax credits, assistance in negotiating the approval process, and, most importantly, exclusive license to market the product for a specific indication for a period of 7 years. Patent protection also offers market exclusivity, but the period of exclusivity under the ODA does not begin until the drug receives final approval from the U.S. Food and Drug Administration (FDA), which may occur years after filing for a patent. Furthermore, the ODA can confer market exclusivity even when a patent is not or cannot be awarded. We are grateful for helpful advice from Norman Barton, Alain Enthoven, Michael Gluck, Alison TauntonRigby, and Judith Wagner. Their assistance should in no way imply their approval. Alan Garber is an HSR&D Senior Research Associate of the Department of Veterans Affairs and a Henry J. Kaiser Family Foundation Faculty Scholar in General Internal Medicine. Ann Clarke is supported by Fonds de la Recherche en Sant6 du Quebec.

583

Goldman, Clarke, and Garber

Alglucerase, a new drug for Gaucher's disease, is precisely the kind of treatment that the ODA was designed to promote. Gaucher's disease is a serious but rare genetic disorder that results from insufficient activity of the enzyme glucocerebrosidase. Government research conducted from the mid-1970s to the early 1980s led to the discovery of alglucerase, an apparently efficacious chemical derivative of the missing enzyme. Large-scale production techniques were necessary to make alglucerase generally available to Gaucher's patients, which became feasible only with its commercial development. It is unlikely that product development could have proceeded as quickly without designation as an orphan. In this paper, we describe the development of Ceredase™, a trademarked version of alglucerase, and the role of orphan legislation in this process. We then discuss the implications of the ODA for therapeutic innovation, the availability of treatment, and health care costs. The case of alglucerase dramatically illustrates the challenges that the United States faces in devising policies to control health care costs without deterring the development of efficacious new technologies. BACKGROUND

Gaucher's disease is an inherited metabolic disorder characterized by the accumulation of compounds called glycolipids, substances ordinarily broken down by the enzyme glucocerebrosidase (2;6;7;8;12;32). A set of genetic defects in Gaucher's patients causes the decreased activity or absence of this enzyme. Consequently, glycolipids, which are primarily derived from white blood cells, accumulate in the spleen, liver, bone marrow, and other organs. The clinical manifestations of glycolipid deposition include abdominal distension, low blood counts, and severe bone pain. Less frequently, the glycolipids can accumulate in the brain, lungs, heart, kidney, and skin. Clinical Heterogeneity

The clinical severity of Gaucher's disease varies dramatically, but the severity may not be closely associated with the specific genetic defect. Several genetic mutations cause this enzyme deficiency (22;31;34;56;57), but according to several researchers, the same mutation may lead to heterogenous presentations, and patients with similar clinical manifestations do not necessarily have the same genetic abnormality (2). Others claim that there is a much stronger association between genetic abnormality (genotype) and disease severity (phenotype) (8;61). A synthesis of the literature suggests there is indeed a link between genotype and phenotype, but its strength is influenced by environmental factors and other unknown conditions. Gaucher's disease is classified into three categories, based on its clinical presentation. Type 1, the adult form, is usually the least severe. Although the name suggests that this condition is present only in adults, it may manifest itself at any time. This chronic condition affects the spleen, liver, and bone marrow. The enlarged spleen is thought to accumulate and destroy platelets and red and white blood cells. The bone marrow, the normal site of production of blood cells, may be unable to replace the destroyed cells because it is also infiltrated with glycolipid. The patient consequently develops anemia, resulting in fatigue and shortness of breath. A low platelet count causes excessive bruising and bleeding. The macrophages, cells which eliminate debris and other contaminants, become congested with unmetabolized glycolipids and restrict blood flow to the bones, thereby decreasing the oxygen supply and causing the bone pain experienced by many Gaucher's patients. This pain can be debilitating, resulting in frequent confinement to bed. The bones are also more likely to deteriorate and fracture, sometimes leading to deformities that restrict the patient to a wheelchair. 584

INTL. J. OF TECHNOLOGY ASSESSMENT IN HEALTH CARE 8:4, 1992

ODA in the development of Ceredase™

Breathing problems can be severe enough to warrant oxygen therapy. Child development can be affected as well; the onset of puberty may be delayed, and the growth of teeth and bones may be impaired. It was thought that Type 1 disease tended to follow a progressive downhill course, but some investigators now believe that once these patients reach early adulthood, the disease's course stabilizes (8). Often Type 1 disease causes minimal symptoms. Type 2, the infantile form, is the most severe of the three types. It usually becomes apparent before 6 months of age and is fatal within 2 years. The involvement of the central nervous system distinguishes this form of the disease from Type 1. Infants with Type 2 disease have abnormal movements and postures, difficulty handling oral secretions, and seizures. Type 3, the juvenile form, is of intermediate severity. Its manifestations are similar to those of Type 2 and include abnormal eye movements, seizures, and dementia, but it involves the neurological system later in its course.

Incidence Little is known about the prevalence of each type of Gaucher's disease. The most common form is Type 1, which primarily affects Ashkenazi Jews. The other two types, which do not have an apparent ethnic predilection, are rare; their birth incidence is believed to be no greater than 1/50,000 (2). The prevalence of the disease is not known with precision. Most estimates of the disease's prevalence are based on the frequency of the gene that causes the disease in the population, a rate which itself is uncertain. According to some investigators, 7.5% (1/13.3) of the Ashkenazi Jewish population carry a form of the abnormal genes that cause Gaucher's disease (8;42). The disease develops only in persons who have 2 such mutations. Ordinarily, this occurs in one-quarter of the offspring of pairs of carriers of Gaucher's disease. If Ashkenazi Jews married only other Ashkenazi Jews (i.e., there was 100% intramarriage within the Ashkenazi Jewish population), then the birth incidence of the disease would be 1/700(1/13.3 x 1/13.3 x 1/4) within this population. If these assumptions are valid, and affected patients do not have shortened life spans, there are about 10,000 people with Gaucher's disease among the 6 million Jewish people in the United States. If Gaucher's patients have higher mortality rates than the general population, and if the intramarriage rate is less than 100%, the actual prevalence of the disease will be lower. Limited data suggest that there are as many non-Jews as Jews with Type 1 disease in the United States (42). The Jewish people tend to have a genetic mutation that results in a less severe form of the disease. By one estimate, 10,000 Jews have Gaucher's disease, but only about 10-20% warrant medical intervention (42). In contrast, of the estimated 10,000 non-Jews with the disease, about 60-80% may be sufficiently symptomatic to require therapy, implying that about 7,000-8,000 Gaucher's patients in the United States would be candidates for medical intervention. Genzyme asserts that only 3,000 Gaucher's patients in the United States (50% of them Jewish) have severe enough forms of the disease to be candidates for treatment, although there may be as many as 3,000 moderately to severely affected patients outside of the United States (46). Estimates of disease prevalence in the United States, based on these sources and others, are listed in Table 1.

Treatment Prior to the introduction of enzyme replacement therapy, medical care was limited primarily to ameliorating the symptoms of the disease. Patients often had their spleens INTL. J. OF TECHNOLOGY ASSESSMENT IN HEALTH CARE 8:4, 1992

585

1:432-1:2,500 1:40,000-1:100,000 1:40,000-1:100,000

Jews

Non-Jews

1:25,000-1:100,000 1:40,000-1:100,000 1:40,000-1:100,000

Birth incidence

2,800-16,000 70-175 70-175

Jews

2,500-10,000 2,500-6,250 2,500-6,250

Non-Jews

Absolute number affected

Types 2 and 3 may be amenable to enzyme replacement therapy but have not been as thoroughly investigated. Sources: 2;8;36;42;59;60; personal communication with Genzyme Corporation.

a

1 2 3

Disease type

Table 1. Incidence of Gaucher's Disease in the United States

280-3,200 NAa NAa

Jews

1,500-8,000 NAa NAa

Non-Jews

Number warranting enzyme replacement therapy

ODA in the development of Ceredase™

removed, based on the belief that the profoundly enlarged spleen trapped and destroyed the blood cells. This practice fell into disfavor because of concern that the removal of the spleen might cause the unmetabolized glycolipid to accumulate more rapidly in other organs and also make the patient more susceptible to infection. The anemia of Gaucher's disease has been treated with blood transfusions and more recently with erythropoietin, the recombinant form of a naturally occurring hormone that stimulates the production of red blood cells. The bone pain has been alleviated with analgesics, oxygen therapy, and bed rest, and the fractures have been corrected with surgery. Treatment of the underlying enzyme deficiency has been a difficult and risky procedure. The most common approach had been bone marrow transplantation, a process in which the cells of a patient's bone marrow are completely destroyed by radiation and chemotherapy and replaced by marrow from a normal donor (21;29;30;32; 33;35;48;50;54). The transplanted marrow contains a normal gene for glucocerebrosidase, and the new cells that the patient produces following transplantation should contain normal levels of the enzyme. This procedure requires a genetically matched donor; even then it remains a potentially lethal treatment for a typically nonlethal condition. Fewer than 20 such procedures have been performed successfully. Although there were early attempts to transplant the spleen and kidney with the hope that these organs would also serve as a source of the enzyme, these efforts were unsuccessful (27;28). Enzyme Replacement

The lack of success of these therapies for Gaucher's disease led to further efforts to correct the underlying enzyme deficiency by replacing the missing enzyme. Since the mid-1970s, investigators at a number of institutions had been attempting to extract glucocerebrosidase from human tissue. Researchers at the National Institutes of Health (NIH) and Scripps Clinic developed a method to harvest the enzyme from human placental tissue (18;40). However, initial efforts to treat at least 18 patients with the enzyme were largely unsuccessful. Investigators were unable to produce sufficient quantities to administer adequate doses, and it became clear that the native enzyme would need to be modified in order to ensure adequate uptake in the cells in the body that needed it most (5;9;10;13;14;15;26).

During the early 1980s, researchers made significant progress toward overcoming these obstacles (l;16;19;20;23;24;25;39;53;55). They developed a more efficient harvesting procedure; they developed a form of the enzyme that was preferentially taken up by macrophages, the cells that use the enzyme to break down glycolipids; and they devised a more effective dosage protocol (4). In the process, the NIH contracted with the New England Enzyme Center to supply the enzyme in accordance with the NIHdevised protocol. By 1981, the New England Enzyme Center had closed down, and Genzyme Corporation, a fledgling pharmaceutical company, took over the contract to supply the enzyme (46). Figure 1 presents a time line of Genzyme's role in the development process. Over time, alglucerase injections emerged as having therapeutic potential in the treatment of Type 1 Gaucher's patients. However, this large-molecular-weight compound is unable to enter the central nervous system and reverse the neurological damage that is found in other forms of Gaucher's disease. This issue is being studied. Development of Ceredase™

Like any company that develops pharmaceutical products and attempts to bring them to market, Genzyme exposed itself to substantial financial risk. Large fixed costs are a ubiquitous feature of the drug development process, as large expenditures are required for basic research and development, mandatory testing, and investment in capINTL. J. OF TECHNOLOGY ASSESSMENT IN HEALTH CARE 8:4, 1992

587

Goldman, Clarke, and Garber Figure 1. The Development of Ceredase™

Date

December March September March March October February April April November

Milestone 1965 Gaucher's disease first attributed to glucocerebrosidase deficiency. 1974 Enzyme replacement therapy first attempted at the NIH as a treatment for the disease. 1981 Genzyme begins to supply the NIH with the enzyme. 1983 First successful treatment with the modified form of the enzyme is begun at the NIH. 1985 Genzyme receives orphan designation for the modified placentaderived enzyme. 1987 Clinical partnership (Genzyme Clinical Partners, L.P.) established with $10 million to fund R&D. 1988 FDA approves Genzyme's IND application for Ceredase™, its trademarked name for the modified enzyme. 1989 Clinical trials begin at the NIH with 12 Gaucher's patients. 1989 Treatment IND protocol approved by FDA. 1990 Clinical partnership bought out by Genzyme Corporation. 1990 Genzyme files an NDA for Ceredase™. 1991 FDA approves Ceredase™ for the treatment of Gaucher's disease (Type 1). 1991 Genzyme receives orphan designation for a recombinant form of the enzyme.

ital, equipment, and marketing staff (17). As a result, production (marginal) costs are often quite low relative to these start-up costs. Companies make these outlays without any assurances that their products will be sufficiently safe and effective to earn FDA approval. Even with approval, their drugs may not be profitable, for they may face competition from similar drugs and alternative treatments. Treatments for rare disorders are particularly vulnerable, because the market size may not be sufficient to justify these development costs. In these instances, a company may need additional assurance that its product will be profitable to gear up production and fund clinical trials. Recognizing this challenge. Congress passed the ODA to encourage the development of treatments for rare disorders (58). Orphan Designation

Genzyme received orphan designation for its product in March 1985. At the time, orphan status represented the only substantial opportunity for alglucerase production to be profitable. Given the NIH's involvement in the development of alglucerase, Genzyme could not rely on patent protection to ensure an adequate return on its investment. Because the NIH and others had published papers discussing harvesting, purification, and modification strategies, these processes were not patentable.1 Without orphan designation, Genzyme would have had to rely solely upon proprietary information to give it a competitive edge. Given the difficulty in maintaining the secrecy of the manufacturing process, such an advantage would have been a shaky foundation on which to begin the development of a new drug. Therefore, by offering Genzyme an opportunity to exclusively market Ceredase™ for 7 years, orphan designation was the impetus behind Genzyme's quest for FDA approval. Development and Production Costs

After orphan status was granted, Genzyme raised US $10 million through a limited partnership that was set up "to develop, manufacture and sell injectable products in588

INTL. J. OF TECHNOLOGY ASSESSMENT IN HEALTH CARE 8:4, 1992

ODA in the development of Ceredase™

corporating modified forms of the enzyme . . ." (52). Beginning in September 1987, Genzyme Clinical Partners was responsible for conducting the human clinical trials, improving the manufacturing process, and seeking FDA approval for Ceredase™ (52). Genzyme Corporation contributed the rest of the funds that were necessary to further develop and market the product (46). This off-the-balance-sheet partnership did not insulate Genzyme from all financial risk; rather, it represented a way to ensure the company's viability should the Ceredase™ project fail. The partnership was bought out by Genzyme in February 1990 after its initial funding ran out (52). Genzyme claims that it spent approximately $58 million to develop Ceredase™ (46). The costs of establishing facilities to produce and market the enzyme, yielding sufficient quantities of the enzyme for clinical trials, and complying with FDA approval procedures are the largest components of this total. Approximately $10 million of these costs were offset by NIH contracts (41). Genzyme did not receive any of the grant assistance that is available through the Office of Orphan Product Development (47). Manufacturing costs for Ceredase™ appear to be very high compared to the costs of manufacturing other pharmaceutical products. The manufacturing process involves extracting the enzyme from large quantities of human placentae, purifying the extract and inactivating viral and bacterial contaminants, and chemically modifying the extract to expose a sugar compound that is recognized by the macrophage receptor. To produce alglucerase in sufficient quantities to treat just 200 patients for 1 year requires 1.8 million human placentae, an amount equivalent to almost half the annual number of live births in the United States. Between October 1989 and April 1991, Ceredase™ was made available to approximately 70 patients around the country under the FDA's Treatment IND program. This program is designed to facilitate access to experimental drugs for the treatment of otherwise untreatable diseases. Under this arrangement, Genzyme sold the drug to patients who were not enrolled in clinical trials whose physicians agreed to abide by the treatment protocol (46;58). As part of Treatment IND regulations, Genzyme was entitled to charge patients a price sufficient to recover the "costs of manufacture, research, development, and handling of the investigation^ drug" (58). The FDA approved a price of $3.00 per unit (45). Although the cost recovery provision refers to research and development (R&D) costs, it is unlikely to be interpreted to mean that a pharmaceutical manufacturer should be able to recoup all of its fixed costs from the small pool of patients who participate in the Treatment IND program. At the $3.00 price, Genzyme would not have been able to recover its estimated $58 million in fixed costs from these 70 patients. In fact, it is difficult to determine whether this price exceeded the manufacturing costs. After final FDA approval, signifying the end of the Treatment IND program, the market price of Ceredase™ rose to $3.50 per unit (46). The Approval Process

Genzyme received approval for an IND application to begin clinical trials of its modified form of the enzyme in March 1988 (45). On the basis of encouraging results from a 1-year study involving 12 patients that was begun at the NIH in March 1989, Genzyme received approval of its New Drug Application (NDA) for Ceredase™, in April 1991. Although the FDA usually requires randomized clinical trials, such studies often are impractical for rare conditions. In approving Ceredase™, the FDA relied almost exclusively on observational studies. The approval of the NDA was expedited by Subpart E designation as part of a new FDA regulation that is not included in the ODA. This regulatory provision is designed to hasten the approval of clearly efficacious therINTL. J. OF TECHNOLOGY ASSESSMENT IN HEALTH CARE 8:4, 1992

589

Goldman, Clarke, and Garber

apies for severely debilitating or life-threatening diseases (58). Genzyme chose not to receive protocol assistance from the Office of Orphan Product Development, relying instead on more informal channels of communication (46;47). Efficacy

Many uncertainties still surround the use of Ceredase™. Studies have been published by the NIH, Scripps Clinic, and Mt. Sinai Medical Center. The NIH's clinical experience with the chemically modified enzyme actually began in 1983, but much of this early work is unpublished. The FDA considered only NIH data in their regulatory review of Ceredase™, along with data provided by Genzyme. The NIH's first efforts with the macrophage-targeted alglucerase involved 8 patients. These participants received a fixed weekly dose of 189 units (49). Only the smallest of these patients showed any clinical improvement. His allotment was equivalent to a dose of 9-12 units/kg, the highest in the study (41). Since that time, he has continued to receive enzyme replacement therapy with weekly to biweekly doses of 30 units/kg (49). His blood counts improved, the size of his spleen and liver decreased, and his bone is rebuilding. To refine the dosage regimen, NIH researchers administered a single dose of the drug to 21 patients. The doses ranged from 0.6 units/kg to 234 units/kg. Forty-four hours later, they performed a liver biopsy (49). The liver glucocerebroside decreased in 8 out of the 11 patients who received greater than 30 units/kg, whereas it decreased in only 1 of the 10 patients who received a lower dose. However, 6 of these 10 low-dose patients had some degree of structural changes in storage deposits in the liver that were detected by electron microscopy. Interpreting these findings to mean that a minimum dose of 30 units/kg was required, NIH investigators enrolled 4 adults and 8 children with moderate to severe Type 1 Gaucher's disease in a study of a higher-dosage regimen (3;38). This is, to date, the only case series that the NIH has published about their clinical experience with Ceredase™. Upon entry into the study, all patients were anemic, had enlarged livers and spleens, and displayed evidence of bone abnormalities on x rays. The participants received 60 units/kg of alglucerase administered intravenously every 2 weeks. Two severely affected patients received weekly alglucerase infusions. After 9 to 12 months of therapy, all of the participants experienced improvement, and no significant toxicity was observed. The blood counts increased, and the spleen and liver sizes decreased in most patients. Bone pain responded less dramatically. All recipients felt that their quality of life had improved as a result of the enzyme replacement therapy. Researchers at Scripps Clinic administered alglucerase to 4 patients with Type 1 Gaucher's disease (11). They used only one-quarter of the total dose recommended by the NIH, but on a thrice-weekly basis. After 4 to 13 months of therapy, all of the patients experienced a decrease in the size of the liver. The blood counts increased in those patients with abnormal values. Fifteen other patients at Scripps and 10 in Israel also are receiving the lower-dose regime. Their response has been similar to the published results from the Scripps study (8;42). A researcher at Mount Sinai Hospital in New York is treating approximately 30 patients with alglucerase, administered as 30-60 units/kg every 2 weeks, with a maximum duration of 18 months (43). He reports that his patients also have had an improvement in their blood counts and liver and spleen size. In both groups of patients, bone pain has diminished, although there is little radiographic evidence of any improvement of the bone disease. 590

INTL. J. OF TECHNOLOGY ASSESSMENT IN HEALTH CARE 8:4, 1992

ODA in the development of Ceredase1"

It generally is accepted that alglucerase injections are beneficial for those patients with Type 1 Gaucher's disease who have sufficiently low blood counts or enlarged livers and spleens to cause symptoms (3;8;11;43;46). It is of no use for acute problems such as crises of severe bone pain or uncontrollable bleeding. Clinical experience with the treatment has been too limited to assess its impact on mortality. However, Genzyme does describe patients who were severely ill prior to the initiation of enzyme replacement therapy but now enjoy a functional lifestyle (46). Furthermore, investigators disagree on the appropriate dose and dosage interval. Most experts believe that the initial dose that is required to metabolize the glycolipid debris in a moderately to severely ill patient must be about 60 units/kg on a biweekly basis. For those patients with less severe disease, an initial dose as low as 10-30 units/kg may be sufficient (41). Others suspect that smaller doses administered every 2 days are more efficacious (8). There is even greater uncertainly about the appropriate dose for long-term therapy. Genzyme believes that the NIH's liver biopsy data suggest that after about 12 months of high-dose therapy, a maintenance dose of 7.5-15 units/kg will be adequate to control the patient's problems. More recent NIH studies of dosage reductions further support the efficacy of maintenance therapy with a low dose of the enzyme (41), although substantial debate surrounds this issue (42). Finally, it is unclear at what stage in the disease a patient should begin therapy. If treatment begins before the patient becomes symptomatic, it might arrest the development of disease. Such an approach is unfeasible, however, because it currently is impossible to predict accurately who will become symptomatic, and it would be prohibitively expensive to treat all Gaucher's patients to prevent future complications. Cost of Treatment

The costs of treatment with alglucerase depend on the particular dosage regimen, varying with dosage (measured in units/kg), dosing interval (number of doses per year), weight of the patient (in kg), and price per unit of Ceredase™. We consider four different regimens: (a) the regimen of 60 units/kg each week that was used by the NIH investigators with 2 patients; (b) the standard NIH dose of 60 units/kg every 2 weeks; (c) a treatment used at Scripps Clinic of 30 units/kg each month, administered three times weekly; and (d) a maintenance dose, as suggested by Genzyme, of 10 units/kg every 2 weeks. Table 2 shows the annual cost of treating each patient and the total costs of treating 300 (an estimate of the population currently undergoing treatment), 2,500 (a reasonable lower estimate of the number of people who warrant medical intervention in the United States), or 8,000 (an intermediate estimate of the number of people in the United States who warrant medical intervention) (46). Under the initial dosing protocol for a moderately to severely affected patient (treatment [b]), the annual cost of Ceredase™ for a 70-kg individual is US $328,200. This price does not include the outpatient costs of administering the biweekly injections and performing the necessary diagnostic evaluations. These ancillary costs range from $4,800-$13,200 annually (44). WELFARE EFFECTS OF ORPHAN DESIGNATION

Ostensibly, Congress designed orphan legislation to overcome disincentives to invest in new health care technologies. Implicit in the law is a conviction that firms need substantial guarantees of monopoly rents and subsidy of research costs in order to undertake the risk and expense of developing drugs for rare conditions. If high fixed INTL. J. OF TECHNOLOGY ASSESSMENT IN HEALTH CARE 8:4, 1992

591

50 50 50 50

52 26 144 26

60.0 60.0 2.5 10.0 $546,000 $273,000 $63,000 $45,500

One person $273,000,000 $136,500,000 $31,500,000 $22,750,000

Currently treated (500)

Extreme projection (8000) $4,368,000,000 $2,184,000,000 $504,000,000 $364,000,000

Intermediate projection (2500) $1,365,000,000 $682,500,000 $157,500,000 $113,750,000

Cost of treating Gaucher population (US$)

Average weight derived from an estimate that 60% of the population in treatment are adults (70 kg) and 40% are children (30 kg). Sources: 3;11;42; personal communication with Genzyme Corporation.

a

a b c d

Regimen

Weight" (kg)

Frequency (annual)

Dose (units/kg)

Table 2. Annual Cost of Ceredase™ Treatment under 4 Dosing Regimens

ODA in the development of Ceredase™

costs and risks of development make high returns an essential incentive to invest, monopoly pricing may be necessary for companies to make such drugs available. To those who will pay for it—insurers, health maintenance organizations, patients, and government programs —and to those who are interested in health care expenditures, the price of Ceredase™ and its cost of production are subjects of keen interest. The high production costs arise from its intensive manufacturing process and expensive material inputs; it may take as much as 1 ton of placentae to treat a patient for a year (37). Furthermore, the existence of public and private insurance distorts the individual's decision to undergo treatment by breaking the link between payment and consumption of health care, thereby diminishing the price sensitivity of demand. If health insurers agree to reimburse patients for treatment with alglucerase, insured patients (and their physicians) will not be deterred by the cost of the treatment. As a result, this market is characterized by a monopoly in which marginal cost is high and demand is extremely insensitive to price. Under these circumstances, a high price is inevitable. With other pharmaceutical products, the structure of demand is similar, but marginal costs usually are much lower. This may explain why Ceredase™ is expensive relative to other medical technologies. Whether alglucerase injections are worth the cost is a question of paramount concern. The limited experience to date strongly suggests that Ceredase™ reduces the morbidity of Gaucher's disease, but its effects on mortality are essentially unknown. However, even if Gaucher's disease were uniformly and rapidly fatal, and Ceredase™ eliminated all mortality due to the disease, it would cost almost $350,000 for each year of adult life saved. At this price, Ceredase™ is not only one of the most expensive treatments ever introduced, but its cost per quality-adjusted-life-year would be well outside the range of accepted therapies (51). If it does not affect mortality, then any enhancement in the quality of life also comes at great cost. It is possible that further technological innovation will improve the efficiency of the manufacturing process and reduce the cost of treatment (8). Recombinant technology represents the most promising approach, because a recombinant form of the enzyme could be safer and cheaper to produce (8;46). However, the development of recombinant technology often poses greater financial risk to the manufacturer than do conventional chemical purification and synthetic techniques. In the presence of this uncertainty, Genzyme may have devoted its efforts toward a less efficient but more predictable production technique. If so, it is natural to ask whether Genzyme's orphan protection for its conventional product will discourage the development of less costly recombinant technology. Genzyme has continued to pursue this new technology aggressively. In November 1991, the company received orphan designation for a recombinant form of the enzyme for the same indication. In general, substances that are shown to be clinically superior or to have a different molecular structure can gain marketing exclusivity for the same indication (58). Unlike the conventional form of the enzyme, Genzyme's recombinant form will be free of the possibility of viral contamination, and the molecular structure will be slightly different (46). Presumably, orphan status was awarded on this basis. However, other manufacturers may have been deterred from developing recombinant technology if they believed that the original market exclusivity that was awarded to Ceredase™ may have excluded their own recombinant form of the enzyme (42). If so, provisions of the ODA may have deterred the development of a less expensive manufacturing process.

INTL. J. OF TECHNOLOGY ASSESSMENT IN HEALTH CARE 8:4, 1992

593

Goldman, Clarke, and Garber CONCLUSION

The ODA is particularly important for drugs, such as Ceredase™, that do not have patent protection and for which the potential market is limited. By rewarding companies that assume the risks of developing and marketing such treatments, its provisions undoubtedly help patients gain access to potentially efficacious treatments. Perhaps the greatest reward to companies that seek orphan status for their products is the 7-year monopoly that orphan designation confers. When no effective alternative treatment is available, and the condition is rare, health insurers may be unable either to deny coverage for the orphan drug or to challenge its price. Under these conditions, monopolists can charge very high prices. Ceredase™ illustrates just how high those prices can become. When Congress drafted the ODA, it may not have intended to raise health expenditures by encouraging the development of extraordinarily expensive technologies, whose long-term safety and efficacy are largely unstudied. It is inevitable that along with the valuable and cost-effective technologies whose availability can be credited to the Act, there will be others that are not cost-effective. The ODA does not instruct the FDA to consider the cost-effectiveness of orphan technologies, and it is not clear if the FDA, whose responsibilities concern the evaluation of safety and efficacy of drugs and medical devices, should decide whether the benefits are commensurate with the costs as part of the approval process. The monopoly status that is conferred by the drug approval process may well be essential to the development of innovative therapies, but the case of Ceredase™ raises questions about whether the process may go too far. NOTE 1

Genzyme believes it now has sufficiently changed the original processes that were developed at the NIH to obtain patent approval. In the words of a Genzyme vice president, their process bears no resemblance to the original NIH procedure. She claims the original process had a much lower yield per unit of tissue with a higher degree of impurity (61).

REFERENCES 1. Aerts, J. M. F. G., Donker-Koopman, W. E., Murray, G. J., et al. A procedure for the rapid purification in high yields of human glucocerebrosidase using immunoaffinity chromatography with monoclonal antibodies. Analytic Biochemistry, 1986, 154, 655-63. 2. Barranger, J. A., & Ginns, E. I. Glucosylceramide lipidosis: Gaucher disease. In C. R. Scriver, A. L. Beaudet, W. S. Sly, & D. Valle (eds.), The metabolic basis of inherited disease, 6th ed. New York: McGraw-Hill, 1989, 1677-98. 3. Barton, N. W., Brady, R. O., Dambrosia, J. M., et al. Replacement therapy for inherited enzyme deficiency-Macrophage targeted glucocerebrosidase for Gaucher's disease. New England Journal of Medicine, 1991, 324, 1464-70. 4. Barton, N. W., Furbish, F. S., Murray, G. J., et al. Therapeutic response to intravenous infusions of glucocerebrosidase in a patient with Gaucher disease. Proceedings of the National Academy of Sciences U.S.A., 1990, 87, 1913-16. 5. Belchelz, P. E., Crawley, J. C. W., Braidman, I. P., & Gregoriadis, G. Treatment of Gaucher's disease with liposome-entrapped glucocerebrosidase: P-glucosidase. Lancet, 1977, ii, 116-17. 6. Beutler, E. Gaucher disease. Blood Reviews, 1988, 2, 59-70. 7. Beutler, E. Monocyte and macrophage disorders: Lipid storage disorders. In W. J. Williams, E. Beutler, A. J. Erslev, & M. A. Lichtman (eds.), Hematology, 4th ed. New York: McGraw-Hill, 1990, 886-94. 8. Beutler E. Gaucher's disease. New England Journal of Medicine, 1991, 325, 1354-60. 9. Beutler, E., Dale, G. L., Guinto, E., & Kuhl, W. Enzyme replacement therapy in Gaucher's disease: Preliminary clinical trial of a new enzyme preparation. Proceedings of the National Academy of Sciences U.S.A., 1977, 74, 4620-23.

594

INTL. J. OF TECHNOLOGY ASSESSMENT IN HEALTH CARE 8:4, 1992

ODA in the development of Ceredase™ 10. Beutler, E., Dale, G. L., & Kuhl, W. Replacement therapy in Gaucher disease. In R. J. Desnick (ed.), Enzyme replacement therapy in genetic diseases: 2. New York: Alan R. Liss, Inc., 1980, 369-81. 11. Beutler, E., Kay, A., Saven, A., et al. Enzyme replacement therapy for Gaucher disease. Blood, 1991, 78, 1183-89. 12. Brady, R. O., & Barranger, J. A. Glucosylceramide lipidosis: Gaucher's disease. In J. B. Stanbury, J. B. Wyngaarden, D. S. Fredrickson, J. L. Goldstein, & M. S. Brown (eds.), The metabolic basis of inherited disease, 5th ed. New York: McGraw-Hill, 1983, 842-56. 13. Brady, R. O., Barranger, J.-A., Furbish, F. S., et al. Prospects for enzyme replacement therapy in Gaucher disease. In R. J. Desnick, S. Gatt, & G. A. Grabowski (eds.), Gaucher disease: A century of delineation and research. New York: Alan R. Liss, Inc., 1982, 669-80. 14. Brady, R. O., Barranger, J. A., Gal, A. E., et al. Status of enzyme replacement therapy for Gaucher disease. In R. J. Desnick (ed.), Enzyme therapy in genetic diseases: 2. New York: Alan R. Liss, Inc., 1980, 361-68. 15. Brady, R. O., Pentchev, P. G., Gal, A. E., et al. Replacement therapy for inherited enzyme deficiency. Use of purified glucocerebrosidase in Gaucher's disease. New England Journal of Medicine, 1974, 291, 989-93. 16. Choy, F. Y. M. Purification of human placental glucocerebrosidase using a two-step highperformance hydrophobic and gel permeation column chromatography method. Analytic Biochemistry, 1986, 156, 515-20. 17. Comanor, W. S. The political economy of the pharmaceutical industry. Journal of Economic Literature, 1986, 24, 1178-1217. 18. Dale, G. L., & Beutler, E. Enzyme replacement therapy in Gaucher's disease: A rapid highyield method for purification of glucocerebrosidase. Proceedings of the National Academy of Sciences U.S.A., 1976, 73, 4672-74. 19. Dale, G. L., Beutler, E., Fournier, P., et al. Large scale purification of glucocerebrosidase from human placentas. In R J. Desnick (ed.), Enzyme therapy in genetic diseases. 2. New York: Alan R. Liss, Inc., 1980, 33-41. 20. Doebber, T. W., Wu, M. S., Bugianesi, R. L., et al. Enhanced macrophage uptake of synthetically glycosylated human placental glucocerebrosidase. Journal of Biological Chemistry 1982, 257, 2193-99. 21. Erikson, E., Groth, G. C , Mansson, J. E., et al. Clinical and biochemical outcome of marrow transplantation for Gaucher disease of the Norrbottnian type. Ada Paediatrica Scandanavica, 1990, 79, 680-85. 22. Eyal, N., Wilder, S., & Horowitz, M. Prevalent and rare mutations among Gaucher patients. Gene, 1990, 96, 277-83. 23. Furbish, F. S., Blair, H. E., Shiloach, J., et al. Enzyme replacement therapy in Gaucher's disease: Large-scale purification of glucocerebrosidase suitable for human administration. Proceedings of the National Academy of Sciences U.S.A., 1977, 74, 3560-63. 24. Furbish, F. S., Steer, C. J., Krett, N. L., & Barranger, J. A. Uptake and distribution of placental glucocerebrosidase in rat hepatic cells and effects of sequential deglycosylation. Biochimica et Biophysica Ada, 1981, 673, 425-34. 25. Grabowski, G. A., & Dagan, A. Human lysosomal P-glucosidase: Purification by affinity chromatography. Analytic Biochemistry, 1984, 141, 267-79. 26. Gregoriadis, G., Neerunjun, D., Meade, T. W., et al. Experiences after long-term treatment of a Type 1 Gaucher disease patient with liposome-entrapped glucocerebroside: P-glucosidase. In R. J. Desnick (ed.), Enzyme therapy in genetic diseases. 2. New York: Alan R. Liss, Inc., 1980, 383-92. 27. Groth, C. G., Collste, H., Dreborg, S., et al. Attempt at enzyme replacement by organ transplantation; renal transplantation in Gaucher disease. Transplantation Proceedings, 1979, 11, 1218-19. 28. Groth, C. G., Dreborg, S., Oeckerman, P. A., et al. Splenic transplantation in a case of Gaucher's disease. Lancet, 1971, 1260-64. 29. Hobbs, J. R., Shaw, P. J., Hugh-Jones, K., et al. Beneficial effect of pretransplant splenec-

INTL. J. OF TECHNOLOGY ASSESSMENT IN HEALTH CARE 8:4, 1992

595

Goldman, Clarke, and Garber

30. 31.

32. 33. 34. 35. 36. 37. 38. 39.

40. 41. 42. 43. 44. 45. 46. 47. 48. 49.

596

tomy on displacement bone marrow transplantation for Gaucher's syndrome. Lancet, 1987, ii, 1111-15. Kirkpatrick, D. V., Barrios, N. J., Shapira, E., & Humbert, J. R. Treatment of enzyme storage disease with matched and partially matched bone marrow transplantation: The Tulane marrow transplant group experience. Blood, 1990, 76(suppl. 1), AZ184. Kleinschmidt, T., Christomanou, H., & Braunitzer, G. Complete amino-acid sequence and carbohydrate content of the naturally occurring glucosylceramide activator protein (Al activator) absent from a new human Gaucher disease variant. Biological Chemistry HoppeSeyler, 1987, 368, 1571-78. Krivit, W. Lysosomal storage diseases. In R. Hoffman, E. J. Benz, S. J. Shattil, B. Furie, & H. J. Cohen (eds.), Hematology. Basic principles and practice. New York: Churchill Livingstone, 1991, 617-20. Krivit, W. Bone marrow transplantation as treatment for storage diseases: Background review and specific application to Gaucher's disease. Japanese Journal of Inborn Errors of Metabolism, in press. Latham, T. E., Theophilus, B. D. M., Grabowski, G. A., & Smith, F. I. Heterogeneity of mutations in the acid-glucosidase gene of Gaucher disease patients. DNA and Cell Biology, 1991, 10, 15-21. Ludman, M. D., Lipton, J., Sadhev, I., et al. Gaucher disease: Bone marrow transplantation in rapidly progressive disease. American Journal of Human Genetics, 1988, (suppl.), A237. Matoth, Y., Chazan, S., Cnaan, A., et al. Frequency of carriers of chronic (type 1) Gaucher disease in Ashkenazi Jews. American Journal of Medical Genetics, 1987, 27, 561-65. Maugh, T. H., II. Study finds a second genetic link to illness. Los Angeles Times, December 1991. Murray, G. J., Howard, K. D., Richards, S. M., et al. Gaucher's disease: Lack of antibody response in 12 patients following repeated intravenous infusions of mannose terminal glucocerebrosidase. Journal of Immunological Methods, 1991, 137, 113-20. Murray, G. J., Youle, R. J., Gandy, S. E., et al. Purification of P-glucocerebrosidase by preparative-scale high-performance liquid chromatography: The use of ethylene glycolcontaining buffers for chromatography of hydrophobic glycoprotein enzymes. Analytic Biochemistry, 1985, 147, 301-10. Pentchev, P. G., Brady, R. O., Hibbert, S. R., et al. Isolation and characterization of glucocerebrosidase from human placental tissue. Journal of Biological Chemistry, 1973, 248, 5256-61. Personal communication with Norman Barton, MD, PhD, Developmental and Metabolic Branch, National Institute of Neurological Disorders and Stroke, U.S. National Institutes of Health. Personal communication with Ernest Beutler, MD, Department of Molecular and Experimental Medicine, Scripps Research Institute, La Jolla, CA. Personal communication with Gregory Grabowski, MD, Department of Genetics and Molecular Biology, Mt. Sinai Medical Center, NY. Personal communication with Glenn Molyneaux, MD, Assistant Medical Director, Blue Shield of California. Personal communication with John Short, MD, Executive Secretary, U.S. Food and Drug Administration Endocrinologic and Metabolic Drugs Advisory Committee. Personal communication with Alison Taunton-Rigby, PhD, Vice President Therapeutics, Genzyme Corporation. Personal communication with Peter Vaccari, Office of Orphan Product Development, U.S. Food and Drug Administration. Rappeport, J. M., Barranger, J. A., & Ginns, E. I. Bone marrow transplantation in Gaucher disease. Birth Defects, 1986, 22, 101-09. Report by Norman Barton, MD, PhD in Proceedings from the Endocrinologic and Metabolic Advisory Committee, October 22, 1990, 135-248.

INTL. J. OF TECHNOLOGY ASSESSMENT IN HEALTH CARE 8:4, 1992

ODA in the development of Ceredase1"

50. Ringden, O., Groth, C. G., Aschan, J., et al. Bone marrow transplantation for metabolic disorders at Huddinge hospital. Transplantation Proceedings, 1990, 22, 198-202. 51. Russell, L. B. Some of the tough decisions required by a national health plan. Science, 1989, 240, 892-96. 52. Securities and Exchange Commission. 10-K filings on behalf of Genzyme Corporation, 1987-1990. 53. Stahl, P. D., Rodman, J. S., Miller, M. J., & Schlesinger, P. H. Evidence for receptor-mediated binding of glycoproteins, glycoconjugates, and lysosomal glycosidases by alveolar macrophages. Proceedings of the National Academy of Sciences U.S.A. 1978, 75, 1399-403. 54. Starer, E, Sargent, J. D., & Hobbs, J. R. Regression of the radiological changes of Gaucher's disease following bone marrow transplantation. British Journal of Radiology, 1987, 60, 1189-95. 55. Strasberg, P. M., Lowden, J. A., & Mahuran, D. Purification of glucosylceramidase by affinity chromatography. Canadian Journal of Biochemistry, 1982, 60, 1025-31. 56. Tsuji, S., Choudary, P. V., Martin, B. M., et al. A mutation in the human glucocerebrosidase gene in neuronopathic Gaucher's disease. New England Journal of Medicine, 1987, 316, 570-75. 57. Tsuji, S., Martin, B. M., Barranger, J. A., et al. Genetic heterogeneity in type 1 Gaucher disease: Multiple genotypes in Ashkenazic and non-Ashkenazic individuals. Proceedings of the National Academy of Sciences, U.S.A., 1988, 85, 2349-52. 58. United States Congressional Office of Technology Assessment Report on the Pharmaceutical Industry. In press. 59. Zimran, A., Gelbert, T., Westwood, B., et al. High frequency of the common Jewish mutation for Type 1 Gaucher disease among the Ashkenazi Jewish population. Blood, 1990, 76 (suppl.), A199. 60. Zimran, A., Gelbart, T., Westwood, B., et al. High frequency of the Gaucher disease mutation at nucleotide 1226 among Ashkenazi Jews. American Journal of Human Genetics, 1991, 49, 855-59. 61. Zimran, A., Gross, E., West, C, et al. Prediction of severity of Gaucher's disease by identification of mutations at DNA level. Lancet, 1989, ii, 349-52.

INTL. J. OF TECHNOLOGY ASSESSMENT IN HEALTH CARE 8:4, 1992

597