Budget impact analysis of orphan drugs in Belgium: estimates from 2008 to 2013

Abstract

Objective: This article aims to calculate the impact of orphan drugs on the Belgian drug budget in 2008 and to forecast its impact over the following 5 years.

Method: The 2008 budget impact was calculated by triangulating information derived from multiple Belgian data sources. The 2008–2013 budget impact analysis was based on three scenarios reflecting different levels of growth in the number of registered orphan drugs in the European Union, the number of drugs reimbursed in Belgium, and the average annual cost per patient per drug in Belgium.

Results: The orphan drug budget impact amounted to €66.2 million (or 5% of the Belgian hospital drug budget) in 2008. The impact would increase to €130–204 million in 2013, depending on the scenario.

Conclusions: This static analysis measured orphan drug costs only, assuming that other components of health expenditure do not change over time. The analysis showed that the budget impact of orphan drugs in Belgium is substantial and rising, thereby putting pressure on total drug expenditure. Policy options to address the rising budget impact include pricing linked to return on investment, risk-sharing arrangements and re-appraisal of orphan drug status if additional indications are approved.

• Belgium
• budget impact
• orphan drugs
• price
• rare diseases
• reimbursement

Introduction

A rare disease is a disease with a very low prevalence. In the European Union (EU), rare diseases are defined as life-threatening or chronically debilitating diseases that have a prevalence of 50 out of 100,000 individuals or less[1]. Due to their low prevalence, rare diseases and their treatment with orphan drugs have traditionally been neglected by industry and by the scientific, medical and political communities[2]. The assertion that orphan drugs target few patients can be questioned in certain cases. For instance, certain orphan drugs have proved to be effective against multiple rare diseases and, thus, target a larger number of patients. An example is sorafenib, which has been approved for the treatment of hepatocellular carcinoma and advanced renal cell carcinoma. Also, mainstream drugs (on and off patent) exist to treat rare diseases. In general, a number of approaches are available to treat rare diseases (e.g. orphan drugs, surgery, nutrition). Nevertheless, with 50 orphan drugs for the treatment of 42 diseases on the European market in March 2010[3], only a small part of the treatment need for rare diseases is covered.

To stimulate the development of orphan drugs, the EU has implemented specific policies in 2000¹. The European Commission may grant orphan designation (i.e., the award of orphan status to a drug) and marketing authorisation (i.e., the approval to market the drug) to a drug, taking into account the quality, safety and efficacy of the drug. Pricing and reimbursement decisions of orphan drugs are a member state responsibility and may be based on factors such as cost effectiveness and budget impact. The pricing of orphan drugs is not different to that of non-orphan drugs. The market conditions are, however, different to those surrounding non-orphan drugs. Legislation in essence creates small virtual monopolies for each orphan drug. As a result, industry has an incentive to ask for high prices. Also, as the costs of research and development have to be recouped from a small number of patients, prices of orphan drugs tend to be high[4]. Member states have few market mechanisms in place to put downward pressure on prices[5],[6].

The reimbursement of orphan drugs and the associated budget impact is challenging the general principles that underpin reimbursement policy in many member states[7]. Due to the small number of patients with a rare disease, economic evaluations of orphan drugs are often hampered by limited evidence on clinical effectiveness. Some European countries apply arbitrary thresholds that specify the maximum cost per unit of outcome that a health care payer is willing to pay for a drug (e.g., £30,000 or €34,000 per quality-adjusted life-year in England and Wales). Economic evaluations tend to find that orphan drugs are not cost effective and do not meet such thresholds because the incremental cost for the additional health benefit provided by the orphan drug is usually high[8]. The budget impact of orphan drugs implies that limited drug expenditure is allocated to few patients, which may challenge the boundaries of social solidarity[9].

In 2008, the Belgian Health Care Knowledge Centre launched a study on rare diseases and orphan drugs given that their specific features make them an issue of high priority for policy makers[3]. The aim of this article is to calculate the budget impact of orphan drugs in Belgium in 2008 and to forecast how this budget impact will evolve over the following 5 years. The findings may serve to raise awareness among policy and decision makers about the budget impact of orphan drugs and to inform policy surrounding orphan drugs.

Patients and methods

The analysis quantifies the budget impact of orphan drugs on the Belgian drug budget in 2008. The 2008 budget impact is then used as the starting point to forecast the budget impact for the 2008–2013 period based on a number of scenarios.

Budget impact: 2008

At the end of 2008, 31 different orphan drugs were approved for reimbursement in Belgium (Figure 1). These 31 drugs corresponded to 35 different indications. The 31 drugs were fully reimbursed by the National Institute for Health and Disability Insurance (the Belgian third-party payer). To estimate the budget impact of orphan drugs, information was combined from several Belgian data sources.

Figure 1. Number of orphan drugs reimbursed in Belgium, 1999–2008.

The research team accessed the ministerial decrees that grant initial reimbursement to a drug in Belgium[10]. These decrees include estimates of the number of patients, the cost per patient per year (based on the defined daily dose and the price per daily dose) and the budget impact. These estimates are based on the original file submitted by the sponsor to obtain reimbursement. Estimates proposed by the industry are reviewed during the reimbursement procedure, and sometimes adapted by the Drug Reimbursement Committee. Estimates reported in the ministerial decree suffer from a number of limitations. The number of patients is usually unknown and analyses that simply apply prevalence figures systematically over-estimate the budget impact because in practice not all patients will be treated with a specific orphan drug. In most cases, some time is needed to actually identify patients who may use the drug: the uptake of the drug is likely to evolve over time. Furthermore, not all patients actually consume the doses as recommended by industry. To account for these aspects of data uncertainty, various other sources were accessed.

The research team drew on figures published by the National Institute for Health and Disability Insurance based on their internal information. These figures were made public at a hearing at the Federal Parliament in February 2009 and were published in the MORSE report[11]. This data source included information about the number of orphan drugs that filed a demand for approval of reimbursement. Additionally, figures were obtained from the Ministry of Economic Affairs. This ministry is in charge of approving drug prices and collects turnover information from the industry on an annual basis. Furthermore, the analysis gathered data from IMS Health.

Finally, information was extracted for those orphan drugs that have submitted a revised file. Orphan drugs need to submit a revised file to the Drug Reimbursement Committee after they have been on the market for 3 years or because they have applied for an extension (e.g., a new dose). These files and the published ministerial decrees reported budget impact estimates based on actual prescriptions during the period that the product has been on the market.

The budget impact estimates were calculated based on the number of patients that are treated (linked to the prevalence) and the average cost of treating a patient.

Finally, using the threshold value used in England of €34,000 per quality-adjusted life-year and the estimated budget impact, a simple analysis calculated the total number of quality-adjusted life-years that an orphan drug would need to generate for all treated patients over the course of 1 year in order to satisfy the threshold.

Budget impact: 2008–2013

Three scenarios were developed to estimate the future budget impact of orphan drugs in Belgium during 2008–2013: a low-growth scenario, a medium-growth scenario and a high-growth scenario. These scenarios were based on the following variables: (a) an estimate of the average annual number of orphan drugs that obtain a marketing authorisation in the EU; (b) an estimate of the annual number of drugs that obtain a positive reimbursement decision in Belgium; and (c) the average annual cost per patient per drug in Belgium.

With respect to the first variable, 48 orphan drugs had obtained a marketing authorisation by the end of 2008. The number of marketing authorisations can be forecasted by examining the evolution of the number of orphan drug designations granted by the European Commission based on the advice of the Committee for Orphan Medicinal Products of the European Medicines Agency (Table 1)[12]. A total of 593 designations were granted by the end of 2008. The analysis assumed that there would be an average increase in the number of marketing authorisations of ten drugs per year. This figure also corresponds to expert opinions and expectations from the European Medicines Agency. Therefore, the number of marketing authorisations by scenario is as follows:

• Low-growth scenario: increase of eight new orphan drugs per year;

• Medium-growth scenario: increase of ten new orphan drugs per year;

• High-growth scenario: increase of 12 new orphan drugs per year.

Table 1.  Overview of European orphan drug designations, 2000–2008.

Year	Applications submitted	Positive COMP opinions	Applications withdrawn	Final negative COMP opinions	Designations granted by European Commission
2008	119	86	31	1	73
2007	125	97	19	1	98
2006	104	81	20	2	80
2005	118	88	30	0	88
2004	108	75	22	4	72
2003	87	54	41	1	55
2002	80	43	30	3	49
2001	83	64	27	1	64
2000	72	26	6	0	14
Total	896	614	226	13	593

COMP = Committee for Orphan Medicinal Products.

To date, 90% of orphan drugs that have a marketing authorisation have gained reimbursement in Belgium (i.e., transfer ratio of 90%)[3]. Therefore, the transfer ratio by scenario is as follows:

• Low-growth scenario: transfer ratio of 80%;

• Medium-growth scenario: transfer ratio of 90%;

• High-growth scenario: transfer ratio of 100%.

The average cost of a reimbursed orphan drug in 2008 was estimated at €2.135 million. On the one hand, the 2008 average cost may not be representative of the annual cost of an orphan drug and may in fact be too low because more than one in three orphan drugs were introduced during the course of the year and therefore generated a cost for a partial rather than for a full year. On the other hand, the 2008 average cost was influenced by a few drugs with a high budget impact and is therefore probably too high. Many drugs are expected to have budget impacts well below that average. Therefore, the average annual cost by scenario was as follows:

• Low-growth scenario: average cost of €2.0 million per drug per year;

• Medium-growth scenario: average cost of €2.135 million per drug per year;

• High-growth scenario: average cost of €2.3 million per drug per year.

Results

Budget impact: 2008

Table 2 presents data on orphan drugs reimbursed in Belgium, their indication, whether their clinical evidence was published in the international peer-reviewed literature, whether a treatment alternative was available, and their supplier. For around 65% of orphan drugs, clinical evidence had been published and alternative treatments existed. No orphan drug was removed from the Belgian market due to adverse events or other reasons.

Table 2.  Characteristics of reimbursed orphan drugs in Belgium.

Orphan drug	Clinical evidence published?	Alternative treatment available?	Supplier	Indication
Aldurazyme (laronidase)	No	No	Genzyme	Mucopolysaccharidosis, type I
Atriance (nelarabine)	Yes	No	GSK	Acute lymphoblastic lymphoma
Busilvex (busulfan)	Yes	Yes	Pierre Fabre Médicament	Hematopoietic cell transplantation
Carbaglu (carglumic acid)	No	No	Orphan Europe	N-acetylglutamate synthetase deficiency
Duodopa (co-careldopa)	No	Yes	Solvay Pharma	Parkinson's disease
Elaprase (idursulfase)	Yes	No	Shire	Mucopolysaccharidosis, type II
Evoltra (clofarabine)	Yes	Yes	Genzyme	Acute lymphoblastic leukaemia
Exjade (deferasirox)	Yes	Yes	Novartis	Chronic iron overload requiring chelation therapy
Fabrazyme (agalsidase beta)	Yes	Yes	Genzyme	Fabry disease
Glivec (imatinib mesilate)	Yes	No	Novartis	Chronic myeloid leukaemia, malignant gastrointestinal stromal tumours
Increlex (mecasermin)	Yes	Yes	Ipsen Pharma	Treatment of primary insulin-like growth factor-1 deficiency due to molecular or genetic defects (primary growth hormone insensitivity syndrome)
Lysodren (mitotane)	No	No	Laboratoire HRA Pharma	Adrenal cortical carcinoma
Myozyme (alglucosidase alfa)	No	No	Genzyme	Glycogen Storage Disease type II (Pompe's disease)
Naglazyme (galsulfase)	Yes	Yes	BioMarin	Mucopolysaccharidosis, type VI
Nexavar (sorafenib)	Yes	Yes	Bayer Healthcare	Renal cell and hepatocellular carcinoma
Orfadin (nitisinone)	No	Yes	Swedish Orphan International	Tyrosinaemia type I
Replagal (agalsidase alfa)	No	Yes	Shire EGT	Fabry disease
Revatio (sildenafil)	Yes	Yes	Pfizer	Pulmonary arterial hypertension
Revlimid (lenalidomide)	No	Yes	Celgene	Multiple myeloma
Savene (dexrazoxane)	No	No	Topotarget	Anthracycline extravasations
Somavert (pegvisomant)	Yes	No	Pfizer	Acromegaly
Sprycel (dasatinib)	Yes	Yes	Bristol-Myers	Chronic myeloid leukaemia
Sutent (sunitinib)	Yes	Yes	Pfizer	GIST, RCC
Tasigna (nilotinib)	Yes	Yes	Novartis	Chronic myeloid leukaemia
Thelin (sitaxentan sodium)	No	Yes	Encysive	Pulmonary arterial hypertension
Torisel (temsirolimus)	Yes	Yes	Wyeth	Renal cell carcinoma
Tracleer (bosentan)	No	Yes	Actelion	Pulmonary arterial hypertension, chronic thromboembolic pulmonary hypertension
Trisenox (arsenic trioxide)	Yes	No	Cephalon	Acute promyelocytic leukaemia
Xagrid (anagrelide)	Yes	No	Shire Pharmaceutical	Essential thrombocytosis
Zavesca (miglustat)	Yes	Yes	Actelion	Gaucher disease

GIST = gastro-intestinal stromal tumour; RCC = renal cell carcinoma.

Table 3.  Estimated budget impact of orphan drugs in Belgium in 2008.

Orphan drug	Reimbursement date	Number of patients treated	Cost/patient/year (€ thousands)	Budget impact (€ thousands)	Number of QALYs required to satisfy €34,000 per QALY threshold
Aldurazyme (laronidase)	1/8/04	9	312	3600	106
Atriance (nelarabine)	1/6/08	24	23 for adult; 4 for child	160	5
Busilvex (busulfan)	1/10/08	44	4.6	205	6
Carbaglu (carglumic acid)	1/9/06		14–1085	1106	32
Duodopa (co-careldopa)	1/3/07	73	49	4000	118
Elaprase (idursulfase)	1/1/08	8	300	1600	47
Evoltra (clofarabine)	1/7/08	10	64	200	6
Exjade (deferasirox)	1/8/07	1080	varies	3000	88
Fabrazyme (agalsidase beta)	1/8/04	48	195	7500	221
Glivec (imatinib mesilate)	1/7/03	120	31–48	4800	141
Increlex (mecasermin)	1/8/08	1	40	40	1
Lysodren (mitotane)	1/1/08	36	6.5	167	5
Myozyme (alglucosidase alfa)	1/5/07	23–33	176 for child; 411 for adult	7800	229
Naglazyme (galsulfase)	1/12/08	0	376	0
Nexavar (sorafenib)	1/4/07	215	24 weeks: 15	3700	109
			30 weeks: 19
			52 weeks: 33
Orfadin (nitisinone)	1/7/06	14	50–100	1200	35
Replagal (agalsidase alfa)	1/8/04	Cost already counted in agalsidase beta
Revatio (sildenafil)	1/6/07	70–142	7–26	1500	44
Revlimid (lenalidomide)	1/4/08		60	5500	162
Savene (dexrazoxane)	1/9/07	29	10	150	4
Somavert (pegvisomant)	1/4/04	70		1600	47
Sprycel (dasatinib)	1/9/07	85–900		4800	141
Sutent (sunitinib)	1/4/07	GIST: 73	GIST: 6	3000	88
		RCC: 180–240	RCC: 21
Tasigna (nilotinib)	1/9/08	Cost counted with dasatinib
Thelin (sitaxentan sodium)	1/1/08	50	32	3900	115
Torisel (temsirolimus)	1/12/08	0	30	0
Tracleer (bosentan)	1/8/04	358	37	4700	138
Trisenox (arsenic trioxide)	1/11/05	4	37	275	8
Xagrid (anagrelide)	1/11/05	320	7	1500	44
Zavesca (miglustat)	1/9/05	2	93	200	6
Total				66,203

QALY = quality-adjusted life-year.

Table 3 shows the number of patients treated, the cost per patient per year, the estimated budget impact, and the number of quality-adjusted life-years needed to satisfy the threshold value of €34,000 per quality-adjusted life-year. In 2008, the impact of orphan drugs on the Belgian drug budget was estimated to amount to €66.2 million. This corresponded to: over 5% of the hospital drug budget; 1.6% of the hospital budget; 1.9% of pharmaceutical expenditure; and 0.3% of health expenditure in Belgium in 2008. The total number of quality-adjusted life-years needed varied between 1 for mecasermin and 229 for alglucosidase alfa.

Budget impact: 2008–2013

Figure 2 presents the forecasted budget impact of orphan drugs in Belgium during 2008–2013 based on three scenarios. The application of the medium-growth scenario would lead to a budget impact of €162 million in 2013 or an increase of 145% over 5 years. This would represent close to 4% of the cost of all drug reimbursements in Belgium and over 10% of the Belgian hospital drug budget. In the low-growth and high-growth scenarios, the budget impact would amount to €130 million and €204 million in 2013, respectively.

Figure 2. Estimated budget impact of orphan drugs in Belgium, 2008–2013.

Discussion

Globally speaking, the orphan drug legislation implemented by the EU can be considered to be a success. Indeed, the number of orphan drug designations and marketing authorisations granted by the European Medicines Agency increased from 182 designations and 14 authorisations by the end of 2003 to 642 designations and 50 authorisations in March 2010. This means that more orphan drugs are becoming available that target life-threatening diseases for which there may be no alternative therapy and that vulnerable patient groups have access to treatment. However, the price of this success is the substantial and rising budget impact of orphan drugs.

The substantial and rising budget impact originates from the high average price of orphan drugs and the steady increase in the number of orphan drugs entering the market[3]. In Belgium, there is no specific budget ceiling for orphan drugs, but the total cost of all reimbursed drugs does have a ceiling. When the global ceiling is reached, there are mechanisms to compensate for over-expenditure. Two-thirds of any over-expenditure is paid for by the industry and the remainder by the National Institute for Health and Disability Insurance. The budget increase of orphan drugs therefore puts pressure on the global ceiling, implying that some non-orphan-drug companies are likely to have to pay for the rising expenditure on orphan drugs.

Different options might be considered by the National Institute for Health and Disability Insurance to contain the budget impact of orphan drugs. First, the Ministry of Economic Affairs (which is responsible for drug pricing) could require a justification for the price of an orphan drug based on detailed information on the investments made and the potential return at a global level. Within the context of multiple treatment approaches to diseases, a price justification would need to be imposed not only for orphan drugs, but also for other drugs and for treatment approaches not involving pharmacotherapy. Also, this approach necessitates that the Ministry of Economic Affairs makes a subjective judgment about an appropriate level of return on investment.

Second, for some orphan drugs (and indeed for other drugs), risk-sharing arrangements between the drug company and the third-party payer, based on price for performance or conditional reimbursement, could be considered[13]. Such arrangements require that effectiveness measures are readily available within the normal treatment regimen and that information systems are in place for following up patients. Furthermore, the minimally expected level of effectiveness on specific endpoints should be clearly specified when risk-sharing arrangements are envisaged. Risk-sharing arrangements could be combined with patient or disease registries. Setting up registries is part of EU policy and is an action line in rare disease plans that many member states have in place or are setting up. Typical purposes of registries are: to describe the natural history of a disease; to facilitate research; to assess safety, effectiveness and cost effectiveness of treatments.

Third, pharmaceutical companies could be required to inform the Drug Reimbursement Committee if and when an additional indication for an orphan drug is granted by the European Commission. If an additional indication is approved, the status of orphan drug could be revisited and the budget impact of the drug for all indications should be reported.

The analysis suffered from several limitations. When calculating the budget impact of orphan drugs in 2008, one orphan drug could not be included in the analysis because of absence of information on the number of patients, the cost per patient per year and the budget impact (i.e., thalidomide). The reader should also note that the budget impact reflects drug costs only and not those of total treatment. Also, the analysis did not include orphan drugs reimbursed through the Belgian Special Solidarity Fund because this fund covers orphan drugs that are not part of the ‘official list’ of reimbursed drugs. The Special Solidarity Fund may grant reimbursement to individual patients after a product has been authorised to enter the market but before reimbursement is approved by the Drug Reimbursement Committee. In 2007, the budget impact of drugs reimbursed by the Special Solidarity Fund amounted to €4 million (with Myozyme (alglucosidase alfa) accounting for €3.5 million).

In light of data uncertainties, the 2008 budget impact was calculated by cross-checking and validating information derived from multiple data sources. To make the analysis less static, the 2008–2013 budget impact analysis was based on three scenarios reflecting different levels of growth in the number of drugs that gain marketing authorisation in the EU, the number of drugs that gain reimbursement in Belgium, and the average annual cost per patient per drug in Belgium. Nevertheless, the analysis focused on the budget impact of orphan drugs only, assuming that other components of health expenditure do not change over time. For instance, the impact of technological advances and ageing populations on the budget impact of orphan drugs was not taken into account. Also, the impact of any price decreases of orphan drugs that could be imposed by the third-party payer in the future was not considered. As a result, the findings give an idea of the order of magnitude of the budget impact of orphan drugs, but do not represent exact estimates.

Future research needs to be carried out to replicate and validate these estimates. Also, orphan drug development is continuing and the orphan drug market is constantly evolving. As a result, some assumptions are likely to change in the future. For instance, the analysis assumed that new orphan drugs will gain marketing authorisation, but did not consider the fact that orphan drugs might be taken off the market. Market exit can be expected to happen in the longer term, for example when new therapies are introduced that replace existing orphan drugs.

Conclusion

Policy makers in many countries are concerned about the budget impact of orphan drugs, although no comprehensive analysis of budget impact estimates has been published to date. This study has shown that the budget impact of orphan drugs in Belgium is substantial and is expected to rise significantly in the future. Although these estimates were specific to Belgium, the finding of a substantial and rising budget impact of orphan drugs is likely to apply to other countries.

Transparency

Declaration of funding: Financial support for this research was received from the Belgian Health Care Knowledge Centre, a state-funded research institution.

Declaration of financial/other interests: The authors have disclosed that they have no conflicts of interest that are directly relevant to the content of this manuscript.