Paediatric investigation plans and waivers

A paediatric investigation plan sets out how a medicine will be studied in children, and in the European Union an application to authorise a new medicine has to come with an agreed plan or a waiver. The Paediatric Committee decides on the plan, and compliance with it is enforced. Preparing a plan turns on what the plan has to cover and when, how waivers and deferrals are justified, how the application is structured and written, and how the procedure with the committee runs, alongside the paediatric study plan that the United States asks of the same product. The requirement has its costs for companies and its results for children's medicines.

What the plan is and why it is required

A paediatric investigation plan is a development plan that sets out the studies and measures a company proposes for its medicine in children, so that the data needed to support the medicine's authorisation for paediatric patients are obtained. ^{1, 2} In it, the company describes the efforts it will make to test the drug in children. ³ The company prepares it during the clinical development of a new investigational medicinal product, and the broad aim of the requirement is that appropriate paediatric studies are performed, so that treating children with innovative products is no longer largely a matter of guesswork. ⁴ A plan should provide enough data to assess the quality, safety and efficacy of the product in children, and with them its benefit-risk profile in the paediatric population. ¹

The requirement comes from EU Regulation 1901/2006, the Paediatric Regulation, which came into force on 26 January 2007. ^{5, 6, 7, 8} A marketing-authorisation application for a new medicinal product has to include the results of studies carried out according to a plan agreed in advance with the Paediatric Committee of the European Medicines Agency, unless the medicine is exempt through a waiver or a deferral. ^{5, 8, 2, 9} The same applies to an authorised, patented product when a new indication, pharmaceutical form or route of administration is sought. ^{5, 8} Because the regulation also stipulates that a marketing-authorisation application should not be delayed by a paediatric development programme, it provides for deferring or waiving some or all of the paediatric measures. ⁸ Generic products are exempt from the requirement, and orphan products are not. ⁷

Pharmaceutical companies had not really tested drugs in children, for fear of side effects and problems, and regulators recognised that there were not enough medicines on the market for children and that testing in children was necessary to offer them more treatment options. ³ The regulation's objective is to improve the health of children in Europe by facilitating the development and availability of age-appropriate medicines and by increasing the information available on the use of medicines in children. ⁶ It encourages manufacturers to research and develop medicines for children's specific therapeutic needs through a system of rewards and through an obligation to plan the development of a medicine for children, for instance by integrating it into the development for adults, and to submit a corresponding plan. ² The plan is the regulation's main tool for ensuring that unmet therapeutic needs in children are researched and appropriate medicines developed. ⁶ The approval of a suitable plan is a necessary first step towards the regulation's goal of more authorised medicines for children. ¹⁰

The regulation is intended to ensure that children are considered within the development of drugs for adults. ¹¹ That is relatively straightforward where a disease exists in both adults and children, and more complex where it does not but the drug might have another use in children, as with ibuprofen for persistent arterial duct in newborns. ¹¹ Drug development targeted specifically at children would require a different framework. ¹¹

The Paediatric Committee

The Paediatric Committee decides on plans, on waivers, which release the applicant from development in children, and on deferrals, which allow studies to be carried out later. ⁷ It is composed of 33 members and 33 alternates: each member state is represented by one member and one alternate, and further members represent the Committee for Medicinal Products for Human Use, paediatric healthcare professionals' organisations and patient organisations. ⁷ The Committee for Medicinal Products for Human Use approves new drugs, but the Paediatric Committee can block a submission, and the agency will not validate a submission without an agreed plan. ⁷

What the plan covers and when it is submitted

The plan summarises the relevant background on the disease and the drug and uses it to justify a paediatric development programme that covers the entire paediatric population. ¹² It must cover every paediatric age group as defined by the ICH E11 guideline. ⁷ Depending on the type of drug and the relevance of the disease to children, specific quality, safety or efficacy measures may be proposed for all or part of the paediatric population. ¹² The plan must include plans for preclinical testing, such as juvenile animal studies, for formulations, such as intravenous forms for preterm newborns and liquids for infants and young children, for the clinical pharmacology that establishes dosing, and for the clinical trials. ⁷ A detailed timetable has to be provided for all the measures proposed, and adhered to. ¹²

Preparing a plan means addressing six core questions. ¹

1. Is there a need for the candidate medicinal product in children?
2. If there is, in which condition should paediatric development occur, considering the proposed indication in adults?
3. In which age groups or paediatric subsets should the development take place?
4. Should there be an adapted formulation and a specific non-clinical package?
5. What clinical measures should the plan contain?
6. Should any measures in the plan, mainly the clinical trials in children, be deferred?

¹

Within the EU and US frameworks, the essential questions a developer answers before the plan is written are these: does the targeted disease exist in children, and from which age onwards; are the mechanisms of drug and disease comparable enough between adults and children to allow extrapolation of efficacy; which dose should be used at which age; what clinical trials are needed; what special formulations are needed; and are studies in juvenile animals of value. ¹³ The company team will know the adult disease but will need additional external expertise for children, and the best approach is to reach tentative conclusions within the company and then challenge them face to face with a group of external paediatric specialists, after which the plan is written. ¹³ A scientifically sound approach makes it easier to convince the regulatory authorities. ¹³ The committee also asks for paediatric development in rare and very rare diseases. ¹³

The plan is submitted early in development, and the sources set the point in different terms. ^{7, 9, 14} One places it at the end of human pharmacokinetics, which the agency interprets as the end of phase 1, before proof of concept. ⁷ Another places it, in principle, before basic phase I studies in adults are complete, which may require an innovative and creative approach to drafting the documents. ⁹ A third places it as early as possible, preferably once adult pharmacokinetic data are available, which is before phase II. ¹⁴

Waivers, class waivers and deferrals

A waiver releases the applicant from the obligation to carry out paediatric measures, that is, clinical or non-clinical studies or the testing of an age-appropriate formulation. ^{15, 16} A waiver is the route where the company does not believe that paediatric development is necessary or feasible. ⁴ The committee can grant one where the condition does not occur in children or in some subsets of the paediatric population, where the product is expected to be unsafe or ineffective in children or in subsets, and where the product appears to offer no significant benefit over existing treatments for the same condition. ¹⁵ The ground that the condition does not occur means almost literally zero cases, not just an extremely low incidence or prevalence. ¹⁴ A waiver will in some instances be granted where the condition does not occur in children, as with Parkinson's disease, and otherwise a series of studies is required to investigate the quality, safety and efficacy of the drug in children so that the appropriate dosing regimens can be chosen. ¹⁷

A waiver may be partial, for specific subsets of the paediatric population, or total, for all paediatric subsets in a given condition, and any request for one has to be thoroughly justified with supporting evidence. ^{15, 12} The committee grants a waiver only when the applicant makes a convincing case that paediatric measures are not warranted, and an applicant's lack of interest in a development programme for some or all of the paediatric population is not an acceptable reason for one. ¹⁶

The agency publishes class waivers, typically for drugs that treat diseases occurring only in adults or where there is reason to believe that the class of drug is unlikely to have adequate efficacy or safety in children. ¹⁶ Where one applies, the applicant can refer to it in the rationale for not conducting paediatric studies in some or all age categories, or for specific indications. ¹⁶ The sources differ on what a class waiver spares the applicant. ^{16, 7} One holds that a class waiver does not absolve the applicant from submitting a plan, even when it covers every age category. ¹⁶ Another states that an application has to come with an approved plan unless the agency confirms in writing that a class waiver applies. ⁷ The agency's decisions on plans and waivers, including modifications, are published and searchable by condition. ¹⁸ Its product-specific decisions may give insight, from similar types of drugs, into whether a waiver may be appropriate for the applicant's drug. ¹⁶

A deferral is the instrument that avoids a delay in marketing authorisation in adults. ¹⁹ Under a deferral, paediatric measures are proposed, but their completion and reporting are delayed, generally relative to the timing of the adult application. ¹⁶ In many but not all cases, paediatric studies can or should be performed after studies in adults have confirmed the product's activity and safety, so a deferral of the initiation or completion of one or more studies in children may be requested and agreed by the committee. ¹⁹ Where the paediatric programme cannot be completed before the adult application is submitted, a deferral may be proposed, and it too must be justified. ¹² A deferral is product-specific and is generally driven by practical considerations, such as the availability of an age-appropriate formulation, the need for further non-clinical studies, or the requirements of a global clinical development strategy, for instance the availability of data from other regions. ¹⁶

Guidance, forms and templates

The final guideline on the structure and content of a plan was published in September 2008. ²⁰ At first sight it is not a particularly user-friendly document: for example, it often refers to specific articles of the regulation that are themselves sometimes hard to interpret. ²⁰ The guidance was also sometimes contradictory about the exact content and structure of the scientific part, sections B to E, and its header numbering mixed letters with Arabic and Roman numerals. ⁴ Companies for the most part wanted to produce compliant plans, but the way to do so was not always clear, and what the committee is looking for is not always easy to guess. ⁴

Alongside the guideline, the agency publishes an electronic form for the plan application and request for waiver, a PDF file sometimes called the plan template. ²⁰ A common misconception is that text for the whole application can be inserted into it, whereas the file is a dynamic form covering Part A of the application, the administrative details of the applicant, the drug and the intended indication. ²⁰ At its end is a table of contents giving the recommended high-level structure for the scientific part of the application, Parts B to F, from which applicants can build their own Word template. ²⁰ That structure is not identical to the organisation of the guideline, but the difference lies only in the order in which information is given. ²⁰

There was no official plan template, and most applicants used the agency's summary report template with internal guidance text, which agency reviewers use to write the assessment reports that the committee then uses to review the application. ^{21, 20} Because that template tells reviewers what to assess and comment on, a writer who addresses those points while authoring can tailor the plan to the committee's expectations and potentially reduce the number of questions raised at review. ²⁰ In February 2013 the agency published a preformatted plan template on its website. ²² Its headings are largely equivalent to those of the previous guidance, but it gives the exact structure the committee expects to see, and slightly more detailed guidance on the content expected under each heading. ²² The agency invites applicants to use it, which one experienced writer takes as an instruction to do so. ²² The template should help dispel some, but by no means all, of the doubts and inconsistencies about the plan document. ²³

The European Commission's guideline on the format and content of plan and waiver applications and on the compliance check contains all the information on what an application needs to contain, and is fundamental reading for anyone preparing one. ¹⁸ The procedural aspects are in the agency's questions-and-answers document, which collects the technical and regulatory issues that have arisen most often in dealings with applicants, and whose questions 6, 8 and 9 are particularly relevant to preparing the application. ¹⁸ It is a must-read, particularly before a question is addressed to the agency, since in most cases the question will already have been answered there. ¹⁸ The agency's website holds several resources for applicants, in particular in its two paediatric sections, and among them a presentation on the resources available when developing a plan or waiver application. ^{18, 10} Scientific guidelines can be found with the agency's search engine and in preselected lists of guidelines of paediatric interest, among which the draft guideline on the pharmaceutical development of medicines for paediatric use and the guideline on the investigation of medicinal products in the term and preterm neonate are particularly relevant. ¹⁸ The agency also periodically organises expert groups on topics relevant to the development of paediatric medicines, and publishes their presentations and outcomes. ¹⁸

In January 2015 the agency published revised documentation on plans to reflect the changes and simplifications the European Commission had introduced in its guideline on the format and content of applications for agreement or modification of a plan and requests for waivers or deferrals, and on the operation of the compliance check and the criteria for assessing significant studies. ²⁴ The revised documents cover the procedures for submitting plan and waiver applications, for re-examination and for the compliance check. ²⁴

Structure of the application

The core deliverable of an application is its scientific part, a document structured according to the agency's guideline. ^{8, 12} The 2008 guidance, in version 3.0.0 of the application form, recommends a high-level structure with the administrative Part A in the electronic form and the scientific part in Parts B to F. ^{25, 20}

• Part B, the overall development of the product, including information on the target diseases and conditions: B.1, the similarities and differences, with B.1.1 on the disease or condition between populations, including prevalence and incidence, and B.1.2 on the pharmacological rationale, covering structure, absorption, pharmacokinetics, pharmacodynamics, metabolism, elimination, mechanism of action and the similarities and differences of the safety and efficacy profile; B.2, current methods of diagnosis, prevention or treatment in paediatric populations; and B.3, significant therapeutic benefit or fulfilment of therapeutic needs.
• Part C, applications for product-specific waivers: an overview of the requests and the grounds for them, based on lack of efficacy or safety, on the condition not occurring in the specified subsets, or on lack of significant therapeutic benefit.
• Part D, the plan itself: the existing data and the overall strategy proposed, covering the plan's indication, the selected paediatric subsets and the existing quality, non-clinical and clinical data; the quality aspects, with the strategy and an outline of each planned or ongoing study and step in the pharmaceutical development; the non-clinical and the clinical aspects, each with the strategy, an overall summary table and a synopsis or protocol outline of each planned or ongoing study; and the timeline of measures.
• Part E, the request for deferrals, and Part F, the annexes.

²⁶

One rendering of this structure numbers the Part D subsections with Roman numerals, D.I to D.V, and another with Arabic ones, D.1 to D.5, under the same content. ^{26, 27} In practice the high-level structure has to be augmented with subsections tailored to the specific application. ²⁰

Team, resources and timeline

Realistic planning of resources and timelines is crucial to preparing a plan on time, and writing one almost always takes more resources and time than first estimated. ²⁵ The overrun can be partly due to a lack of experience, and partly because a plan often needs a substantial amount of text drafted anew rather than adapted from existing material: background on the paediatric population and the disease, which can be hard to obtain, and the rationale for the development strategy, which typically goes through lengthy discussion and multiple revisions. ²⁵ Many applicants have little or no experience of preparing a plan, and depending on the drug and the disease it can be a demanding document to write, with the effort often underestimated. ²⁸ A recommended length of below 50 pages sounds brief, and can lead sponsors preparing their first plan to underestimate the time and effort required. ⁸

The resources depend on the company's size and structure, but in general the core team should include at least a regulatory coordinator and an experienced medical writer, together with representatives of the chemistry-manufacturing-and-controls, non-clinical and clinical functions, and a publisher to compile the submission package. ²⁹ The functions contribute material for their own areas, including text offered as ready to be just added to the document, which experienced writers know invariably needs extensive reworking. ²⁹ The writer's input keeps the content consistent between sections, and the writer's oversight keeps track of what material is still needed, and by when, if the submission date is to hold. ²⁹ An experienced writer can help the applicant interpret the guideline so that team members can supply the material and scientific guidance the sections need. ²⁸ The main challenges for the writer are applying the guidance to the drug and disease at hand and obtaining the right input from the project team. ¹² By interacting proactively with the team, the writer's goal should be to make the application as clear and focused as possible, so that only a minimal number of issues are raised during the committee's review. ²⁸

The submission date of the plan is usually linked to one of the committee's monthly meetings, in the context of the expected date of the marketing-authorisation application, by which a compliant plan is required. ²⁹ Planning back from that application, it is prudent to allow for the questions that will follow the initial submission, and realistically it can take at least six months between submitting the plan and obtaining agreement on the measures in it. ²⁹ The time needed to prepare the plan depends on the resources available and the extent of information to include, and even for a well-resourced project around six months is realistic for a submission-ready plan, once literature searches, advice on the paediatric strategy, authoring, at least two rounds of review, finalisation and compilation, including the Part F annexes, are counted. ²⁹ The time from starting to prepare a plan to obtaining the committee's agreement can therefore easily extend to a year, and sometimes longer. ²⁹

Part B: the background to the strategy

In Part B the applicant provides the background that supports the rationale for the paediatric strategy proposed later in Parts C and D. ³⁰ Part B.1, which gives information on the disease to be treated and the expected performance of the drug, is generally the hardest part to write, and in it the guideline asks for the known and expected similarities and differences between adult and paediatric populations, and between age categories within the paediatric population. ³⁰ Its topics include the characteristics and seriousness of the disease, its prognosis and epidemiology, the drug's pharmacological properties and mechanism of action, and known or expected safety and efficacy information related to that mechanism. ³⁰ Phase I clinical pharmacology data belong in Part B.1, and safety and efficacy data from studies in adults are summarised in Part D. ³⁰

Because the paediatric population is highly diverse, the difficulty writers often encounter is obtaining information for Part B.1 that covers every age category. ³⁰ Disease characteristics and epidemiology often require a literature search, which can be time-consuming and therefore expensive. ³⁰ It is not unusual for a writer to be told at a kick-off meeting that all the information is in the investigator's brochure, which is rarely, if ever, the case. ³⁰ The writer is then often left to educate the team about the section, the purpose it serves and, depending on the indication, the often considerable effort needed to research the information and summarise it at the level a plan needs. ³⁰

Part B.2 requires a review of current methods of diagnosis, prevention and treatment of the disease in paediatric populations, which may again need a literature search, along with consultation of regulatory information and drug approvals. ³¹ It summarises the current treatments and standard-of-care options across the whole paediatric population and how the applicant's drug compares with them, and where applicable the same information justifies the choice of active comparators in the proposed studies. ³¹ Against the background of Parts B.1 and B.2, Part B.3 justifies the anticipated therapeutic benefit of the drug. ³¹ The key question is whether the drug is expected to give improved safety or efficacy compared with current options in some or all of the paediatric population, or comparable efficacy or safety with an improvement in quality of life, for instance through a better dosing regimen or an age-appropriate mode of administration. ³¹

The paediatric strategy and its measures

Supported by Part B, the applicant has to crystallise a paediatric strategy the committee will accept. ¹⁶ A primary aim of the regulation is that clinical studies be designed and conducted to produce paediatric data that can serve as the basis for the drug's prescribing recommendations, but depending on the drug, the disease and its epidemiology across the paediatric population, the applicant may instead propose a waiver or a deferral for measures in some or all of the population. ¹⁶ A common situation is that the applicant proposes measures for at least part of the paediatric population and a waiver for the remainder, and the writer then has to craft a convincing rationale for the design of those measures that harmonises with the background in Part B. ¹⁶ That is not trivial, because applicants often tend to propose a minimal number of measures, which may be hard to align with Part B and may also contradict the committee's somewhat academic view of the need for paediatric measures. ¹⁶ The rationales for waivers in Part C and for the proposed measures in Part D therefore usually go through a high degree of iteration between the writer and the rest of the team over several weeks before everyone agrees them, and the same is true, to a lesser extent, of the rationale for a deferral in Part E. ¹⁶

Part D, the core of the plan, opens with summaries of existing information on the drug's formulation and on its non-clinical and clinical development, including an overview of the efficacy and safety data from studies in adults. ¹⁶ Clinical pharmacology data do not belong there, because they will already have been summarised in Part B.1 in the context of the drug's pharmacological properties and mechanism of action, and Part D cross-refers back to Part B where necessary. ¹⁶ The proposed measures, whether quality measures such as the development of an age-appropriate formulation or non-clinical and clinical studies, have to be specific in strategy, method description and timelines, and the agency provides a PDF template for synopses of non-clinical and clinical studies. ¹⁶ Here the writer is often confronted with the team's wish to be as non-committal as possible, but anecdotal evidence consistently suggests that the committee requires specific descriptions of measures that may be conducted several years in the future, including details of statistical analyses. ¹⁶ In general a company tries to limit the scope of its commitments as far as possible, particularly as the exact direction of future clinical development is hard to predict and it wants to avoid studies that add no value to its product. ⁴ The timelines of all the measures are specified relative to the submission date of the marketing-authorisation application, to at least the nearest quarter year, and are binding: once the plan is agreed, they can be changed only through the laborious procedure of an amendment. ¹⁶ The details of the measures and their timelines are usually the subject of intense discussion within the team, and the writer can play an important role in aligning the measures across functional areas and with the background on the drug and the disease in Part B. ¹⁶

The key binding elements forms, one of the main novelties of the February 2013 template, set out what the applicant is actually committing to, and are used to check compliance with the plan during and after approval. ²² They contain less information than the synopsis forms that came before them, which suits companies, and they will probably make the plan easier to write, because the actual commitments are often what generates most discussion within a team. ²² The clinical strategy still has to be carefully thought out and presented in the plan. ²²

Justifying the proposals

Every request and proposal in an application, for a plan, a deferral, a waiver or a specific study, has to be properly justified. ³² The committee has viewed several applications negatively, not because the proposals were unacceptable in principle, but because they were not properly justified, which left too few elements to assess whether they were acceptable. ³² A well-written plan is central to a rapid validation and increases the chances of a positive opinion. ¹⁰ Trials in paediatric medicines may be small, but the evidence still needs to be good. ¹⁰

The putative mechanism of action of the product has to be described in sufficient detail, because the committee identifies the condition for paediatric development partly from the mechanism of action, starting from the proposed indication in adults. ³³ Pharmaceutical forms and quality aspects are given in the application form, Part A, rather than in the scientific documents, Parts B to E, again in sufficient detail. ³⁴ Most plans contain at least partial extrapolation, in the sense that the scale of the development, such as the number of studies and patients, differs from that in adults, which rests on the assumption of at least some similarity in response between adults and children with the same or an analogous disease. ³⁵ Whatever the degree of extrapolation, and in some cases complete extrapolation of efficacy may be acceptable, the applicant should be explicit in justifying the amount proposed. ³⁵ The agency has a working group on extrapolation, whose outcomes were to be published on its website from 2012. ³⁵ In some situations, where a product is being developed for more than one condition in adults and the marketing-authorisation procedures will be separate, it is convenient to request two plans for the same product, one per condition, which may, under certain circumstances, allow an earlier reward. ³⁶

The agency accepts requests for presubmission meetings from prospective applicants, with a view to improving the quality of the application and increasing the chances of a smoother validation and a positive opinion from the committee. ³⁷ A draft application can be discussed there for further advice. ¹⁰ The details of these meetings are in question 26 of the agency's questions-and-answers document. ³⁷

Frequent mistakes

Some mistakes and misunderstandings seem to occur in applications with greater frequency than others. ³⁸

• Insufficient information. Too little information, whether in the application form, Part A, or in the scientific document, Parts B to E, is likely to lead to non-validation.
• Excess information. There is no need for a textbook-length discussion of a common disorder.
• Justification. Sufficient justification is crucial, particularly for a waiver or a deferral: it is not enough to state that a disorder is rare in children and that studies are therefore not feasible, and such a request should rest on a prevalence analysis supported by available literature evidence and expert opinion.
• Deferral. Deferral is a frequent source of confusion. A deferral of, say, the completion of a given study means only that marketing authorisation in adults can be sought before that study is complete in children, and it does not exempt the applicant from proposing justified and sufficiently detailed elements of how the study will be run. A deferred study must also carry a proposed completion date given as an absolute date, not relative to the foreseen date of the marketing-authorisation application, and after that date the study becomes due even if the adult application has been postponed.
• Pharmaceutical form. Not all quality aspects of the paediatric product will be known at the time of the application, but the applicant still has to propose what will be developed in enough detail for the committee to evaluate the proposal.
• Non-clinical development. The non-clinical section should discuss explicitly whether studies in juvenile animals are needed, and the committee's non-clinical working group assesses every relevant proposal.
• Clinical studies. The opinion the committee adopts will not contain the full details of each study protocol, only the key elements against which compliance is later checked. Applicants surprised by such a slim opinion often ask for the missing elements to be reinstated, which is unnecessary and can be counterproductive, because a modification of the agreed plan may then become necessary to change secondary elements of the protocol.
• Coverage of all paediatric subsets. Every paediatric subset must be covered in the application, either by studies or by a partial waiver, and a common mistake is to omit a small subset, say from 4 to 6 years of age. In principle, wherever there is a paediatric need a waiver is inappropriate, and the subset must be covered by one or more studies, which may include extrapolation studies.
• Methodology. A single-arm, open-label study cannot demonstrate efficacy, and at best supports a claim of activity, usually on a biomarker or surrogate endpoint; such studies are not always unacceptable, but they need proper justification. A commonly encountered omission is the lack of a power analysis or sample-size determination, again without justification. Rare conditions are even rarer in children and fully powered, controlled efficacy studies are not always possible, but that does not necessarily mean a waiver will be granted, since limited data on activity and tolerability or safety in small samples may be acceptable in some circumstances, again with proper justification. Other common mistakes are multiple primary endpoints without the correct methodological approach, too wide a delta for a non-inferiority study, and mixing too many objectives, such as phases I, II and III, in one study.

³⁸

Completing the package

Even once the team has agreed the content of Parts B to E, time has to be planned to bring the package to submission readiness. ³⁹ With the content stable, the writer finalises formatting, cross-referencing and consistency of language, and attends to technical requirements the agency specifies, such as removing active hyperlinks to references, tables and figures, and ordering the list of published literature at the end of the plan alphabetically. ³⁹

The annexes in Part F are completed in the same period. ³⁹ Besides copies of all the published literature cited in Parts B to E, they should include other relevant reference material where available, such as an investigator's brochure, a risk management plan, product information if the drug is already approved, any opinions or decisions received for the drug from regulatory authorities, and any official scientific advice received on it. ³⁹ Ideally the annexes are compiled as far as possible while the text is drafted, but in practice much of the work is often done shortly before submission, and careful planning is needed to stop it becoming the rate-limiting step. ³⁹ Electronic compilation of the submission requires a publisher, and in practice the writer also acts as the interface between the publisher and the rest of the team for resolving last-minute issues. ³⁹

The procedure with the committee and compliance

Most key components of a paediatric development plan are negotiated between the authority and the company. ⁴⁰ The plan's life cycle has three phases: preparation; submission and negotiation; and execution and modification. ²¹ Submission begins with a letter of intent two months before the plan is submitted, for which the agency's paediatric website gives a template, and the agency then names three key people: the agency's paediatric coordinator, the committee's rapporteur and the committee's peer reviewer. ²¹ The 20 paediatric coordinators are the procedural and administrative link between applicant and committee, and also help with teleconferences and give procedural advice. ²¹ The plan, the application form, the clinical study form and a cover letter are sent to the coordinator, the rapporteur and the peer reviewer by post or courier and through EudraLink, the European medicines regulatory network's secure file-transfer system, and once the submission is validated the entire documentation goes on CD-ROM and through EudraLink to all committee members and alternates, with the references as PDF files on the CD-ROM. ^{21, 41} To meet deadlines, receipt through EudraLink is accepted. ²¹

The negotiation consists of two 60-day blocks. ²¹ The committee's day 30 discussion is documented in a report, and the day 60 discussion produces a list of requested modifications and starts a clock stop. ²¹ Once the requests have been considered, the applicant sends a response document to the agency together with the modified plan, which the committee discusses at day 90 before issuing a new list of requested modifications. ²¹ The last amendments take place between days 90 and 120, and an oral explanation may be requested at day 120 for remaining questions. ²¹ If the committee is satisfied with the modified plan it issues a positive opinion, and if not, the applicant can choose between a negative opinion, re-examination and withdrawal of the plan. ²¹

From 1 August 2016 the agency transmits the committee's opinions and its subsequent decisions to applicants in electronic format only, as a PDF sent through EudraLink, and applicants no longer receive paper versions unless they opt out to receive hard copies. ⁴¹ Neither the opinions nor the decisions carry a signature. ⁴¹ The date on which the applicant first opens the EudraLink message counts as the date of receipt of the attached document for calculating procedural timelines under Regulation (EC) No 1901/2006, and EudraLink records that date automatically. ⁴¹ EudraLink keeps file attachments for up to 90 days only, so applicants should download and archive the attached documents on receipt, and the agency retains a read-only version in its electronic archives. ⁴¹ The agency presents the change as a response to stakeholder feedback collected over the years, with benefits that include faster delivery, more convenient receipt and a shift towards greener solutions in line with its environmental policy. ⁴¹

Once the plan is agreed, the clinical trials and other measures in children begin on the committed timelines. ²¹ Where an agreed plan is no longer feasible, or becomes inappropriate in the light of new scientific knowledge, the applicant can request one or more modifications to it. ¹ In the agency's estimate, the average plan requires three to five modifications. ²¹

After approval in adults, the plan is checked for compliance, that is, whether the company has done what it said it would, and the approval can be revoked if the company has not complied with its obligations. ⁴ The plan commits the applicant to perform the agreed studies in children, and one experienced writer counts it among recent legislation that seemingly increases the regulatory burden in Europe, adding that it could be debated whether the effort might be better employed in other development activities. ⁴² Whatever its usefulness, the legislation is rigorously enforced: when the agency's decision obliging a company to consider paediatric indications other than those in its adult programme was challenged in the European Court of Justice as a misuse of power, the court upheld the agency's original decision. ⁴² An applicant dissatisfied with the outcome of the procedure can also sue at the European Court of Justice, and by 2012 one company that did so had lost both at first instance and in the main trial. ²¹

The US paediatric study plan

The European regulation parallels US legislation, but radically increases its scope. ⁷ For companies, participation is partially voluntary in the United States and compulsory in Europe. ⁴⁰ In the United States, the FDA Modernization Act of 1997 offered six months of patent extension in return for voluntary drug research in children, and the mandatory Pediatric Research Equity Act complemented it in 2003, both laws having to be re-authorised every five years, with the next re-authorisation due in summer 2012. ¹¹ The 1997 legislation encouraged the generation of some paediatric data, at a time when for many drugs there were none at all. ⁴³ The sources name the 1997 law differently: one names the FDA Modernization Act, and another names the Best Pharmaceuticals for Children Act as the first US initiative, in place since 1997, that the European framework followed. ^{11, 5}

For compliance with the Pediatric Research Equity Act, the FDA requires a paediatric assessment at the end-of-phase-2 meeting and a paediatric plan at submission, which is reviewed during the approval procedure. ²¹ For a reward of six months' patent prolongation, a company can negotiate a written request to investigate the drug in children in another indication. ²¹ The first round of paediatric negotiations is usually with the European committee, and negotiations with the FDA come later. ¹³

The Food and Drug Administration Safety and Innovations Act included provisions strengthening the existing paediatric legislation, the Pediatric Research Equity Act and the Best Pharmaceuticals for Children Act. ¹⁴ Before the Act, a paediatric study plan, the US equivalent of the European plan, could be submitted after the product had been approved in adults, and under it the plan must be submitted at the end of phase II development, within 60 days of the end-of-phase-II meeting, although it can be submitted earlier. ^{14, 23} That is more in line with the European plan, which should be submitted as early as possible. ¹⁴ An early submission gives the agencies a much greater say in the paediatric development, can ensure that the programme is sufficiently detailed and, importantly, can expedite paediatric approval and so shorten the period of off-label use in children, with its greater uncertainties about dosing, efficacy and safety. ¹⁴

The two documents are structured differently, and against the European Parts B to E the US plan has twelve numbered sections. ^{14, 27}

1. An overview of the disease in the paediatric population.
2. An overview of the drug or biological product.
3. An overview of extrapolation to specific paediatric populations.
4. The request for product-specific waivers.
5. A summary table of planned non-clinical and clinical studies.
6. Paediatric formulation development.
7. Non-clinical studies.
8. Additional data to support studies in children.
9. Clinical studies, split into paediatric clinical studies and clinical effectiveness and safety studies.
10. The timeline of the paediatric development plan.
11. The plan to request deferral of paediatric studies.
12. Agreements for paediatric studies with other regulatory authorities.

²⁷

Although the European structure appears much more complex, the two plans should be of similar length if both adhere strictly to their guidance: the agency's questions and answers suggest a maximum of 50 pages per condition for the European plan, and a US plan should not exceed approximately 60 pages if the section lengths in its template are observed. ¹⁴ Material from one plan can often be slotted into the other, such as material from section B.1.1 of the European plan into section 1 of the US plan, but adaptation will often be necessary, and the epidemiological data in particular need a European focus in the European plan and a US focus in the US plan. ¹⁴

Both regimes allow a waiver for expected lack of efficacy or safety and for lack of significant therapeutic benefit. ^{14, 44} The third European ground is that the disease does not occur in the target population, and the third US ground is that the necessary studies are impossible or highly impracticable, for example because the number of patients is small. ^{14, 44} One experienced writer reads this as making epidemiological arguments of low patient numbers more likely to succeed in a US plan. ⁴⁴ For particular age groups, a US plan can also include partial waivers based on difficulty developing an appropriate paediatric formulation. ⁴⁴ In the United States, a waiver granted on the grounds of expected lack of efficacy or safety must be reflected explicitly in the product label. ⁴⁴

Regulatory affairs departments face managing two paediatric plans with different timelines and somewhat different formats while maintaining an overall coherence in global paediatric development, which may be particularly problematic when the two review procedures overlap. ^{45, 23} The supposed greater dialogue between the agency and the FDA may in principle help limit diverging opinions, but one experienced writer is sure that differences will arise from time to time, given the different document structures and the differences in the underlying legislation. ⁴⁵ Companies are keen to keep the commitments of the two plans fully compatible, to avoid further unnecessary burdens. ^{45, 23} In 2013, both companies and agencies were still feeling their way. ⁴⁵

Costs and results of the regulation

Paediatric drug development is expensive. ⁴⁶ The EU rewards it with six months of patent extension through a supplementary protection certificate, but the extension comes at the end of patent life, while the additional development costs have to be paid earlier. ⁴⁶ One view holds that a company that has developed a drug up to proof of concept will not abandon it because it has to submit a plan, but that a point will come at which companies stop investing in therapeutic areas where the committee's demands turn the potential profitability of future drugs around. ⁴⁶ A possible example is epilepsy, where one new compound ended up with 16 trials or measures across paediatric epilepsy syndromes, neonatal seizures, epilepsy with partial-onset seizures, and idiopathic generalised epilepsy with primary generalised tonic-clonic seizures. ⁴⁶ The agency classified melanoma as a juvenile disease on an incidence of 1.7 per 100,000 in 15- to 19-year-olds, from US Surveillance, Epidemiology, and End Results statistics, but two-fifths of that group are 18 or 19 years old and so adults, and the number of juvenile melanoma patients is probably half the cited figure. ⁴⁶ In 2011 two companies had a melanoma plan approved. ⁴⁶

One experienced writer holds that the pharmaceutical industry was not proactive in the years before the regulation, focusing on requesting a longer extension of the supplementary protection certificate when better solutions could have been found. ⁴⁶ In the same writer's view, each company now faces an established paediatric structure at the agency, and even when requests are unbalanced there is little a company can do. ⁴⁶ Agency representatives announced at a paediatric conference in autumn 2011 that the degree of detail in each plan would be less in future, which would reduce the workload on both sides. ⁴⁶ One proposal is an independent European institute that would work out binding recommendations for each drug, financed by the pharmaceutical industry, with regulators, industry and clinicians on its board, and that would probably cost a fraction of what all companies together had invested in the more than 1000 plans submitted. ⁴⁶ Consideration of children has become an essential part of drug development and will not go away, and for the moment each company has to negotiate a programme from which children will benefit and that allows the company to survive, a workload much of which is borne by medical writers. ⁴⁶ Medical writers are increasingly involved in designing and documenting overall paediatric development plans, such as the European plans, and their individual components, such as protocols and study reports. ⁴⁰ The regulation was due for a first review in 2013 and a second in 2018. ⁴⁶

On 15 November 2016 the European Commission launched a public consultation, based on a report prepared by the agency and its Paediatric Committee, to support its second report on the regulation, which would assess the regulation's impact on public health and the pharmaceutical industry and was expected in 2017. ⁶ Comparing 2004 to 2006, immediately before the regulation came into force, with 2012 to 2014, the number of new medicines and new indications centrally authorised for paediatric use more than doubled, from 31 to 68. ⁶ Among them are medicines for certain rheumatology conditions in children, for infectious diseases such as chronic hepatitis C and HIV infection, for hypertension and for paediatric cancers such as acute lymphoblastic leukaemia, available on the basis of studies conducted in children. ⁶ In November 2016, clinical trials in children initiated as part of an agreed plan represented about 30% of the paediatric trials recorded in the EU clinical trials database, EudraCT, and by the end of 2015 the committee had adopted 860 opinions on plans. ⁶ The European Network for Paediatric Research at the EMA, Enpr-EMA, was set up to facilitate clinical studies in children, and is an umbrella network of 38 national and international networks recognised for their paediatric research experience, acting as a platform for sharing good practice and as a pan-European voice for promoting research into medicines for children. ⁶

In 2017 the Commission published a 10-year report on the implementation of the regulation, which showed an overall success, with an increase in authorised medicines for children, but identified challenges, in particular that certain therapeutic areas, such as oncology and neonatology, still lacked sufficient development for children. ² On the basis of that report, the agency and the Commission held a multi-stakeholder workshop in 2018 on ways to improve the regulation's implementation, and from it developed a paediatric action plan to provide immediate solutions within the existing regulatory framework. ² In February 2023 the action plan's closing report highlighted the initiatives regulators in the EU had taken over the previous four years to make paediatric regulatory processes more efficient and boost the development of medicines for children. ² At that time the Commission was finalising a proposal to revise the EU's pharmaceutical legislation, including the legislative framework for medicines for children, whose adoption was expected the following month. ² Three of the key improvements the action plan brought are these. ²

1. A strengthened focus on unmet medical needs: over four years the agency and stakeholders met systematically to identify the areas where medicines for children are particularly needed and to shift the research agenda to them, in multi-stakeholder strategy fora on the needs of children with cancer and of children with inflammatory bowel disease, involving clinicians, paediatric patients and their representatives, academia, regulators, health technology assessment bodies and developers, and the agency takes what those meetings learned into account when discussing plans for new medicines.
2. Increased alignment of data requirements between decision-makers, through the agency's paediatric cluster with the US Food and Drug Administration and other international regulators, and through Enpr-EMA's work to align international requirements for the authorisation and standards of paediatric clinical trials.
3. Regulatory processes adapted and simplified, a major outcome being a pilot for a stepwise plan agreement, launched with its own guidance in February 2023: in certain cases a partial development programme can be agreed, conditional on a full plan once evidence becomes available over time, so that a plan can be agreed for an innovative medicine where information crucial to defining parts of it is not yet available, with conditions and milestones set for the company to return to the committee and discuss the uncertainties once more data exist.

²

In June 2019 the agency's human medicines committee recommended extending the indication of Victoza (liraglutide) to children and adolescents aged 10 years or older with type 2 diabetes, on the evidence of a placebo-controlled trial in 134 patients aged 10 to 17 years carried out in accordance with a plan agreed by the Paediatric Committee. ⁴⁷