The influence of the European paediatric regulation on marketing authorisation of orphan drugs for children

Background Drug development for rare diseases is challenging, especially when these orphan drugs (OD) are intended for children. In 2007 the EU Paediatric Drug Regulation was enacted to improve the development of high quality and ethically researched medicines for children through the establishment of Paediatric Investigation Plans (PIPs). The effect of the EU Paediatric Drug Regulation on the marketing authorisation (MA) of drugs for children with rare diseases was studied. Methods Data on all designated orphan drugs, their indication, MA, PIPs and indication group (adult or child) were obtained from the European Medicines Agency (EMA). The outcome and duration of the process from orphan drug designation (ODD) to MA, was compared, per indication, by age group. The effect of the Paediatric Drug Regulation, implemented in 2007, on the application process was assessed with survival analysis. Results Eighty-one orphan drugs obtained MA since 2000 and half are authorised for (a subgroup of) children; another 34 are currently undergoing further investigations in children through agreed PIPs. The Paediatric Drug Regulation did not significantly increase the number of ODDs with potential paediatric indications (58% before vs 64% after 2007 of ODDs, p = 0.1) and did not lead to more MAs for ODs with paediatric indications (60% vs 43%, p = 0.22). ODs authorised after 2007 had a longer time to MA than those authorised before 2007 (Hazard ratio (95% CI) 2.80 (1.84-4.28), p < 0.001); potential paediatric use did not influence the time to MA (Hazard ratio (95% CI) 1.14 (0.77-1.70), p = 0.52). Conclusions The EU Paediatric Drug Regulation had a minor impact on development and availability of ODs for children, was associated with a longer time to MA, but ensured the further paediatric development of drugs still off-label to children. The impact of the Paediatric Drug Regulation on research quantity and quality in children through PIPs is not yet clear.


Introduction
Rare diseases are defined as life-threatening or chronically debilitating conditions with such a low prevalence that special combined efforts are needed to ensure adequate medical care. As a guide, a prevalence of less than 5 per 10,000 citizens in the European Union (EU) is considered low [1]. A low prevalence still equals to approximately 250,000 patients in the Community for diseases near the cut-off point. Much rarer diseases only affect a few dozen patients in the whole EU. There are between 5000 and 8000 rare diseases identified so far, affecting an estimated 30 million EU citizens [2]. Over 80% of rare diseases have a genetic background, with the great majority being single-gene defects, although multifactorial and chromosomal defects exist. Other non-genetic rare diseases are due to degenerative and proliferative causes, infectious diseases, treatment-related toxicities, alimentary deficiencies, rare poisonings and injuries [2,3]. Rare diseases can occur at any age but approximately half of these have their onset at birth or during childhood [4].
Drugs for rare diseases are classified as orphan drugs (ODs). Developing ODs is very challenging. This is mainly due to the various factors that limit clinical studies such as the small number of patients, the heterogeneous and scattered populations, ethical issues (i.e. the use of placebo), lack of validated biomarkers and endpoints, poor diagnostics and limited clinical expertise [5], but also by the lack of return of investment in the small target population [6]. To stimulate research, development and placing on the market of ODs, in 2000, incentives were put in place for drug developers, such as a ten-year marketing exclusivity, access to centralised authorisation procedures and fee reductions for regulatory activities (such as protocol assistance, MA applications and inspections) by the European Medicines Agency (EMA) [6]. Another mechanism that may boost the discovery and development of ODs is "repurposing". This refers to the exploitation of known drugs for new indications [7]. Repurposing receives attention in both the United States and Europe, although differences exist between both continents with respect to policies for repurposing of medicinal products [8]. Several initiatives have been created to identify possible targets for drug repositioning [7]. One of them is the US Food and Drug Administration (FDA) Rare Diseases Repurposing Database to encourage repurposing for rare diseases [9]. There are many examples of ODs that were successfully developed from repurposed drugs [10,11].
The development of drugs for children with rare diseases poses even more challenges than it does for adults. The biology of the growing child, its changing physiology and psychology are much different from adults and requires research that is dedicated to children [12]. Such research is confronted by technical difficulties and legal and ethical constraints. As a consequence, there is little or no investment in research and development of drugs for the paediatric population. More than half of medicines used for children were never or incompletely studied in this population; their use in children is either unlicensed or off label, i.e. out of the scope of the drug's authorised label for age, route of administration, dose frequency, formulation or indication [13]. Use of unlicensed drugs or off-label use is especially common for children with rare diseases and is potentially inefficacious and hazardous [14].
The European Regulation (EC) No 1901/ 2006, hereinafter referred to as the 'Paediatric Drug Regulation' [1] came into force on 26 January 2007 with the objective to improve the health of European children by facilitating the development, accessibility and safe use of new drugs for children aged 0 to 17 years, through clinical studies. These objectives should be achieved without subjecting children to unnecessary clinical trials and without delaying the authorisation for other age populations. This regulation obliges applicants to submit study results to the EMA for each new medicine, new indication, new route of administration or new formulation, according to an agreed Paediatric Investigation Plan (PIP). This PIP describes the planned paediatric studies and their timelines. It should ideally cover all age groups from birth to adolescence. Paediatric studies may be (partially) 'waived' if studies are not feasible, appropriate or safe for (a subset of ) the paediatric population or 'deferred' if it is appropriate to conduct studies in adults prior to initiating studies in children or if studies in children will last longer than studies in adults. The PIP should also describe the need for the development of age-appropriate formulations and/or additional non-clinical information (such as developmental toxicity studies in juvenile animal). When the PIP is completed and all requirements are met, applicants are rewarded with a six month extension of patent protection. Off-patent products developed exclusively for use in children are granted eight year data-and ten year market exclusivity for the paediatric indication (the Paediatric Use Marketing Authorisation (PUMA)). ODs are rewarded with two additional years of market exclusivity. The Paediatric Drug Regulation was introduced in stages (see Table 1) distinguishing new medicinal products from already authorised medicinal products.
The implementation of the Paediatric Drug Regulation has paid off for non-orphan medicinal products. Five years after its implementation, more medicines have become available for children and more research has been conducted in children [15]. Over 600 PIPs have been agreed upon and 30% of those PIPs include studies with neonates, the most neglected group. In addition, more paediatric clinical trials were conducted and the proportion of clinical trials including children increased over the last 6 years, to approximately 10% [15].
For ODs, the impact of the Paediatric Drug Regulation has not been studied. In this manuscript, we describe the drug application process from orphan drug designation (ODD) to marketing authorisation (MA) and analyse the effect of the Paediatric Drug Regulation on the success rate and time course of obtaining MA.

Orphan drug designations and marketing authorisations
The EMA kindly provided us with a list of all ODDs from 2000 until December 2012 with designation date and number, indication and age category for which the OD is intended (children and/or adults, i.e. potential paediatric use or not). Designations are issued for treatment indications so that drugs with more than one indication occur more than once in this list. Of all ODDs that obtained MA, the European Public Assessment Reports (EPARs) and the product information (Summary of Product Characteristics [SmPC]) available at the EMA website (www.ema.europa.eu) were studied. The following information was extracted: authorisation date, approval conditions (i.e. conditional approval or under exceptional circumstances), indication and age category (children, adults or both) for which the product was authorised. All authorised ODs were cross referenced with the Community register (available at http://ec.europa.eu/health/documents/ community-register/html/index_en.htm) of both ODs and drugs for human use to determine whether the authorised products still benefited from an orphan status or not.

Paediatric investigation plans
Of all authorised ODs, submitted PIPs including waivers, deferral agreements as well as description and timelines of the required studies (clinical, non-clinical and formulations) were extracted from the EMA website. As a rule, a PIP has to include all subsets of the paediatric population, but waivers for the entire paediatric population (full waivers) or for certain age groups (partial waivers) are granted when one of the following conditions are met: the condition only occurs in the adult population; clinical studies cannot be expected to be of significant therapeutic benefit or are not feasible; the product is considered unsafe or ineffective in children. Full waivers can be granted for classes of medicinal products ('class waivers') or for specific medicinal products ('product specific waiver').
In case of a deferral, the initiation or completion of paediatric studies described in the PIP is postponed until after MA for adults [1].

Time course to marketing authorisation
The influence of the Paediatric Drug Regulation on the time course of obtaining MA after ODD was analysed using SPSS software version 22.0 (IBM Corp, Armonk, NY, USA). Chi-square was used for subgroup analyses and Cox proportional hazards models were used to examine the impact of potential paediatric use (intended for children yes/no) and approval after or before 2007 (the year in which the Paediatric Drug Regulation was introduced) on time to MA for a designated orphan indication. In order to analyse these effects, the database was restructured from a list of drugs with children and adult indications as separate variables to a table with drug-indication-age combinations. A drug with multiple indications for both children and adults can thus appear more than once in this table. ODD after or before 2007 was entered as separate variable. Time to event was computed as the time elapsed between ODD and MA for every drug-indication-age combination. Since a correlation can be assumed between obtaining authorisation for a paediatric indication and an adult indication for the same medicinal compound, the analysis was also repeated for drugs irrespective of indication by age. When designated orphan products were not authorised yet, the case was censored at the date of analysis.
The effect of repurposing could only be analysed for products with MA using the definition as described by Norman (2013) [10]. This variable was not available for drugs without MA (censored cases). Therefore this variable was not analysed with survival analysis but with General Linear Model (GLM) only. The mean time from ODD to MA was also calculated using the GLM procedure with the following covariates: after/before 2007 and paediatric indication (yes/no).
Differences were taken as significant at P < 0.05. *New indications, pharmaceutical forms and/ or routes of administration which are protected by a supplementary protection certificate or by a patent which qualifies for the granting of a supplementary protection certificate. All three categories need to comply with the requirements of Article 7: application for MA should include the results of all studies conducted in compliance with an agreed PIP or a decision of the EMA granting a (partial) waiver or deferral [1].

Orphan drug designations and marketing authorisations
From the implementation of the OD Regulation in 2000 until December 2012, 1088 ODDs were granted, 670 (62%) were intended for children (either exclusively for children (n = 161) or for both children and adults (n = 509)) and 418 (38%) were for adults only ( Figure 1). As of November 2013, 81 of all granted ODDs had obtained MA. Sixty-five of these were identified as having a potential paediatric use at the time of ODD. Forty of these have indeed become available for children ('onlabel'); 25 potential paediatric products were still off label for children at the time of MA and 16 products were for adults only.
Of the 40 on-label paediatric ODs, 16 are currently under further development for a subset of the paediatric population. The PIP details of these ODs are specified in Table 2.
Of the 25 ODs that are authorised for adults but still off label for children, fifteen products are currently undergoing further development for use in children (Table 3), while the remaining ten are not. Two of those were granted a product specific waiver because the medicinal product did not represent a significant therapeutic benefit over existing treatment for paediatric patients while the remaining eight products did not have a PIP (see below for further details).
For the sixteen ODs for adults, four class waivers were granted for the following conditions: multiple myeloma (n = 2), myelofibrosis (n = 1) and chronic lymphocytic leukaemia (n = 1). Five product specific waivers were granted because the condition only occurs in the adult population (n = 3), because the medicinal product did not represent a significant therapeutic benefit over existing treatment for paediatric patients (n = 1) or because the product was likely to be unsafe in the paediatric population (n = 1). For four products, no PIP was found. One of these was authorised after 2007 but before the implementation of article 8 (in January 2009). Noteworthy is the fact that the potential indication of three products was initially considered for 'adults only' at time of ODD, but these products are currently undergoing paediatric investigations, meaning that they are considered to be of potential paediatric use after all (Table 4).

Paediatric investigation plans
For 36 authorised ODs no decision or information about a PIP was found. For the majority of the products a PIP was not required because approval was granted before the Paediatric Drug Regulation came into force (n = 19) or because application for MA was submitted before the implementation of article 7 (n = 4) or article 8 (n = 1). Unless the applicant files for extension or variation of the initial MA, these medicinal products are likely to remain off label to children. The remaining 12 products without a PIP were developed for (a subgroup of ) children.
For 34 authorised ODs, the PIP was required to include development and testing of an age appropriate formulation or conducting non-clinical and clinical studies.       Most of these (30/34) were granted a partial waiver, the remaining four products were required to develop and assess treatment for the complete paediatric population. None of the PIPs were completed at the time of application for MA as some of the requirements in the PIP were deferred. Partial waivers were mostly granted based on the expectation that clinical studies would be of no significant therapeutic benefit or fulfil no therapeutic need of the paediatric population. PIP decisions, waiver conditions and expected date of PIP completion are described in Tables 2, 3

and 4.
Half of the 34 products with a PIP were required to either develop an age-appropriate formulation or to assess the acceptability of the existing formulation ( Table 5). The majority of these measures applied to oral formulations (n = 13). An age-appropriate diluted formulation was required for intravenous (n = 1) and subcutaneous (n = 2) formulations.
For 15 products non-clinical studies had to be performed. The required measures mostly included juvenile animal studies to determine pharmacokinetics, tolerability, toxicology and/ or toxicokinetics. In some cases, specific pharmacology, exploratory or dose ranging studies were required in vitro or in other animal models.
All 34 products with a PIP required at least one clinical study in children (median = 3, range 1-9 studies). A quarter (n = 25) of the studies were randomised double-blind, placebo controlled studies in the target population. Another 20 studies were open label comparative trials and were either dose-comparative or using an active comparator, historical controls or standard care as controls. The majority of studies were, however, uncontrolled or observational "all-in-one" trials gathering as much data as possible in the target paediatric population, including efficacy, safety, tolerability, activity and/or pharmacokinetics (Table 5).
To date, only one orphan medicinal product completed its PIP (Glivec®), all other PIPs are still on-going. For four products the therapeutic indication has been extended to the paediatric population [15,16]. On average, it takes seven years before PIPs are expected to be completed.   Figure 3A). The same effect was observed when multiple indications of one drug were grouped (data not shown). Potential paediatric use did not prolong the overall drug development process compared to 'adults only' medicinal products (Hazard ratio (95% CI) 1.140 (0.767-1.696), p = 0.52, Figure 3B). The mean (95% CI) time to authorisation for paediatric medicinal products after and before 2007 was 4.04 (3.02-5.07) and 2.93 (1.93-3.92). The mean (95% CI) time to authorisation for adults-only products after and before 2007 was 4.45 (3.78-5.12) and 2.07 (1.52-2.63) years.

Time course to marketing authorisation
Repurposing does not provide any benefit in shortening the authorisation process for neither paediatric nor adult ODs (p = 0.21, GLM, data not shown).

Discussion
More than 80 ODs, covering nearly 100 indications, were authorised in Europe since 2000. Half of these products are available for (a subgroup of ) children. Another 34 authorised ODs are currently undergoing further investigations in children. The introduction of the Paediatric Drug Regulation was associated with a longer time to MA for OD, did not significantly increase the number of ODDs with potential paediatric indications and did not lead to more MAs for paediatric indications.
In this study we were able to quantify the time to authorisation and the number of paediatric ODs, but could not extract the quality of research conducted in children given the relatively young EU Paediatric Drug Regulation. The use of Cox regression to analyse time to MA as a survival function is appropriate and the data set is large enough to draw valid conclusions. There is some autocorrelation between indications for children and adults within the same drug. This means that the time to MA for a paediatric indication is linked to that for adult indications of the same drug, because, in part, they share study results. The data set illustrates that ODs often obtain MA for adults first, for which clinical studies are easier to conduct, and later for children. A control group would have been desirable, but since non-ODs do not have the same starting point (time of obtaining ODD), comparison in this context is not possible and data would have to be based on different criteria which is beyond the scope of this study.
Administrative processes are not static, they change over time, and that also applies to the approval of ODs. This implies that the time to MA, modelled as survival time in our Cox regression model, may not be completely independent of time. This time dependency was addressed by using after/before 2007 as a separate categorical variable. Since the granted therapeutic indication at the time of MA is the result of the assessment of the quality, safety and efficacy data submitted with the marketing application, this may be different (narrower) to the indications proposed at the time of ODD application [17]. After 2007, the Paediatric Committee (PDCO) safeguards that for any potential paediatric medicinal product an investigation plan is made.
The situation for patients with rare diseases has, without a doubt, improved dramatically after 2000, the year in which the EU Orphan Drug Regulation was implemented. Before 2000 only eight products, so called orphan-like drugs, were authorised for the treatment of rare diseases with the support of the EMA [18]. Four of these orphan-like drugs were authorised for use in children. In contrast, the United States introduced the Orphan Drug Act almost 20 years earlier, in 1983 [19]. Over the period 2000-2009, 148 (13%) of 1138 ODDs received MA in the U.S., of which 81% were potentially beneficial for children [20]. In the same period in the EU only 55 of 703 of ODDs were authorised and only 52% of the products were authorised in children.
The Paediatric Drug Regulation, implemented to increase the availability of effective and safe drugs of good quality for children, was also beneficial for ODs. The majority of ODs with potential paediatric use that were off label to children at the time of MA (40% of all ODs) is currently in development for the paediatric population. Also, 40% of ODs authorised for children are undergoing further investigations to either expand the intended treatment group to include younger children    **Not contraindicated in children, however the SmPC mentions a special warning (such as "The effects of medicinal product on children and adolescents have not been studied" or "limited information on the use in children"). and/or to develop an age appropriate formulation for youngsters. This would presumably not have taken place without the instalment of the Paediatric Drug Regulation.
A drawback is the high number of deferrals for both ordinary and orphan products. In deferrals, either initiation or completion of paediatric studies is postponed until the medicinal product is authorised for use in adults, to ensure that it is safe to do research in children and that availability for adults is not delayed.
In the 5 year progress report on the paediatric regulation, the EMA concluded that authorisation of medicines for adults was not delayed. However, in our analysis, products authorised before 2007 had a shorter time to MA than those authorised after the Paediatric Drug Regulation came into force. Apparently the Paediatric Drug Regulation added complexity to the R&D and regulatory process of orphan medicinal products, exemplified by the applicants' investments time and effort in drafting a PIP.
Others also expressed concern that the EU Paediatric Drug Regulation retards drug development and authorisation for adults by demanding paediatric trials, especially for rare diseases [21,22]. It is not only the Paediatric Drug Regulation that causes delay. There are potential other product-and company-related factors such as the indication for which a drug is being developed, the type of drug product in development, the company's experience in developing OD and the size of the companies submitting the MA application [23][24][25], and incentives such as those for Small and Medium Sized companies implemented in December 2005 [26]. Other economic and bureaucratic issues such as the increasing amount of regulations where applicants have to comply with during drug development in general have their effects. However, the increased approval time after 2007 can also be an artefact, caused by the submission of ODD applications increasingly earlier in the developmental phase.
Given the relatively young EU Paediatric Drug Regulation, there are few data on PIP completion and outcome in rare diseases, especially since deferrals lead to an additional seven years before expected completion of the files for paediatric indications. So far, only one product successfully reached completion of development. Furthermore, no orphan-designated medicine has yet obtained the orphan incentive of two additional years of market exclusivity. The impact of introducing PIPs will become apparent in the next few years when more PIPs are expected to be completed and will learn whether applicants are compliant with measures and timelines agreed upon in PIPs.
We could not demonstrate that repurposing is an effective strategy for the development of drugs for rare diseases in children. Drug repurposing is considered an interesting acceleration and facilitation of OD development at lower cost and with lower risk of failure, since these drugs have already been studied [7]. Although repurposed drugs have already been studied in animals and/or humans to some extent, a positive benefit/risk balance has to be established for the intended paediatric population. Since research in children on average takes another seven years after safety and efficacy have been confirmed in adults, this is considered to be the ratelimiting step, irrespective of repurposing.
There is a need for novel research tools to support decisions that balance between exposing children to experiments and the obvious need to provide children with authorised good quality drugs. Comparative trials are considered the primary instrument to collect the evidence needed for MA. However, for rare disease this is often not feasible. In most cases, the studies requested in the PIPs were open label uncontrolled studies. Most studies were designed to collect as much data as possible, ranging from pharmacokinetics and dose finding to safety and efficacy. When experimental research is not feasible, on-going data collection through registry/ observational programs (such as named patient programmes (NPP) or compassionate use) are in place to characterise both long-term safety and efficacy as well as to determine patient characteristics and disease progression [27,28]. There are differences in implementation of legislation throughout the EU. For example, the French authorities explicitly mention that investigation is not the goal of an NPP and that an NPP may not replace a clinical trial [29].
Several novel research strategies have been proposed, such as meta-analytic approaches, extrapolation, modelling and simulation. With the use of sparse sampling, population pharmacokinetics/pharmacodynamics (POP-PK/PD) and/or physiologically based pharmacokinetic models, extrapolation from adults to children, interpolation between paediatric age subgroups and the optimal use of scientific literature and in vitro/preclinical data, drug development is enriched while minimising the burden of studies in children [30]. Since the implementation of the Paediatric Drug Regulation, especially simulation and modelling are increasingly used for paediatric drug development [15].
Non-clinical juvenile studies are often used to bridge the knowledge gap between mature and immature systems, to detect safety issues early and to predict the dose in children. A recent survey showed that in the majority of juvenile toxicity studies, findings were comparable to those for adults, yielding no new information [31]. Furthermore, novel toxicity was uncommon and could have been predicted from either known pharmacology or from adult data. On the other hand, in a preliminary review of 5 completed juvenile animal studies required in PIPs, unexpected organ toxicity and increased sensitivity was observed in 3 medicinal products, stressing the importance of conduction juvenile animal studies [15]. It confirms that drug development is not a "one size fits all" process. A case-by-case evaluation process is necessary, especially for paediatric ODs.
Part of paediatric drug development is to avoid duplication and to ensure that ongoing and planned paediatric research is transparent. To this purpose, in March 2011, the EU Clinical Trials Register (EU-CTR) was made publicly accessible (EU-CTR) for paediatric trials included in a PIP [15]. The website (available at https:// www.clinicaltrialsregister.eu) provides public access to information extracted from the EU clinical trials database (EudraCT), such as protocols and known results. The clinical trials included are those with agreed PIPs from investigator sites within and outside the European Economic Area (EEA). As soon as a paediatric trial is approved, it becomes accessible in the database.

Conclusions
The EU Paediatric Drug Regulation did not increase the number of ODDs with potential paediatric indications nor did it lead to more MAs for paediatric indications. It was associated with a longer time to MA for both adult and paediatric orphan indications. Nonetheless, the Paediatric Drug Regulation has ensured the further paediatric development of drugs still off-label to children. The impact on the quality and volume of research in the paediatric population through PIPs will become clear in the coming few years. Case-by-case assessment, based on innovative research tools is necessary to collate the best evidence while protecting children from unnecessary experiments. Competing interests ACE Pharmaceuticals BV specialises in medical need products and orphan drugs. The authors declare that they have no competing interests.
Authors' contributions ARKV and PJdV designed the study, collected data and performed analysis and interpretation of the data. RvdV and AdB were involved in writing and revising the manuscript. All the authors read and approved the final manuscript.
Author details