The development and accessibility of pediatric medicines in the EU are subject to special funding and regulation during the approval process. Following the entry into force of the Pharmaceutical Market Reorganization Act (AMNOG) in 2011, new medicines undergo a benefit assessment in accordance with Section 35a of the Social Code (SGB V) as part of market access. The additional benefit is assessed by the Federal Joint Committee (G-BA), and reimbursement amounts are then negotiated with the National Association of Statutory Health Insurance Funds (GKV-Spitzenverband). Medicines with a new therapeutic indication are also subject to benefit assessment. This includes, among other things, expanded therapeutic indications for pediatric patients. Medicines with a so-called PUMA (Paediatric Use Marketing Authorization) approval are also affected by AMNOG. The following describes the practical experience to date with the AMNOG benefit assessment for medicines with such approvals. The focus here is particularly on considering the special features of conducting pediatric clinical trials.

> Regulation (EC) No. 1901/2006, in force since 2007, is of central importance for the development and approval of paediatric medicines. The regulation aims to promote the development and accessibility of medicinal products for use in the paediatric population. At the same time, it is intended to ensure that medicinal products are developed and approved as part of ethically acceptable and high-quality research. Since July 26, 2008, a paediatric investigation plan (PIP) with a development program for use in children must be submitted for newly approved medicinal products. A PIP must be approved by the Paediatric Committee (PDCO) at the EMA and is part of the marketing authorization documents; deferral or exemption can be requested (BfArM 2023). Another special form of drug marketing authorization under this regulation is the Paediatric Use Marketing Authorization (PUMA). This authorization, limited exclusively to use in the pediatric population, can be granted for off-patent medicines that are already approved for adults. This authorization may also include the development of child-friendly dosage forms.
At the same time, integral recitals of Regulation (EC) No. 1901/2006 state that the above-mentioned objectives should be achieved “without subjecting the paediatric population to unnecessary clinical trials and without delaying the approval of a medicinal product for other age groups.” In this regard, the work of the PDCO must examine “whether the studies have a potentially significant therapeutic benefit for the participating paediatric patients or for the paediatric population in general.” The principle of avoiding unnecessary clinical or other trials with the paediatric population is of central importance here and is therefore also the subject of Article 1 of the regulation. At the same time, it is stated that paediatric clinical trials may require specific expertise and specific methodology.
Of particular relevance in the context of this analysis is the possibility of using pediatric extrapolations during approval. This refers to the process of generating and transferring evidence regarding efficacy and safety when it can be assumed that the course of disease and the expected response to a drug are sufficiently similar in the pediatric and reference populations (adults or older children) (EMA 2017, EMA 2018). Equally relevant in this context is the use of models and simulation techniques for deriving findings on pharmacokinetics and pharmacodynamics.
Framework of the AMNOG benefit assessment for paediatric medicinal products
As already mentioned in the introduction, medicinal products with a PUMA approval are covered by the AMNOG assessment. This is directly regulated in Chapter 5, Section 2, Paragraph 1 of the VerfO. Accordingly, "medicinal products for which an authorization for paediatric use has been granted in accordance with Articles 5 to 15 of Regulation (EC) No. 726/2004 in accordance with Article 38, Paragraph 1 of Regulation (EC) No. 1901/2006" are considered medicinal products with new active ingredients. New indications for pediatric patients approved via a PIP are also subject to AMNOG assessment in accordance with Chapter 5.
Section 2 paragraph 2 of the VerfO of the broad definition for new areas of application and thus also the AMNOG obligation (G-BA 2023).
Since the GKV-Arzneimittelversorgungsstärkungsgesetz (AMVSG) came into force on May 13, 2017, there has only been a legal requirement for the handling of pediatric medicines under Section 5, Paragraph 5a of the AM-NutzenV (Medicinal Products Benefit Assessment Ordinance). This regulates the G-BA's requirement to review the transfer of evidence for PUMA medicines. Accordingly, "the Federal Joint Committee examines whether an additional benefit can be recognized for patient groups or sub-indications covered by the approval, but which are not or not sufficiently represented in the study population and for which approval was granted based on the transfer of evidence. In these cases, it can recognize an additional benefit provided that the transfer of evidence is permissible and justified according to the current state of scientific knowledge, including with regard to the benefit assessment." It further states that the G-BA regulates the details in its rules of procedure. However, when looking at the G-BA's Rules of Procedure, there are no regulations that would describe a procedure for transferring the evidence.
The legal justification for the AMVSG emphasizes that, due to the special requirements for conducting clinical trials with children, it may happen, particularly with PUMA medicinal products, that insufficient evidence is available to demonstrate an additional benefit for all approved patient groups. The requirement to review evidence transfer therefore takes into account the "special features of medicinal products approved for pediatric use," according to the justification. At the same time, it is recognized that "special requirements apply to studies in children and, in principle, the number of study participants must be kept low" (BMG 2016). The limiting wording of the requirements and the justification suggests that this was clearly intended to address the special issue from the 2015 benefit assessment of the PUMA medicinal product propranolol (Hemangiol®). In this assessment, the G-BA extrapolated the results of a decision between different patient groups with varying degrees of disease severity. The G-BA also followed the assessment of the regulatory authority regarding extrapolation, taking into account the specific features and limitations of conducting pediatric clinical trials with regard to avoiding unnecessary trials as well as the specific expertise and methodology that may be required (G-BA 2015).
Although the AM-NutzenV only requires the assessment of evidence transfer for PUMA medicinal products, the G-BA still reserves the right to examine the possibility of evidence transfer for pediatric extensions of the indication. This became clear, among other things, from the benefit assessment of the active ingredient vandetanib on July 6, 2017 (just a few days after the AMVSG came into force) (G-BA 2017). The procedure, with an extension of the indication to children 5 years of age and older in the treatment of thyroid cancer, outlined the possible model for the procedure for evidence transfer.
For the target population of adults, the G-BA previously saw an indication of minor additional benefit. The EMA's findings regarding the identical cause of the disease, the comparability of the clinical picture, comparable pharmacokinetic parameters, and comparable efficacy and safety based on the study data from the randomized controlled adult study and the single-arm pediatric study were used as a minimum requirement for evidence transfer (G-BA 2017). A further key criterion was the appropriate comparator therapy defined by the G-BA as identical for children and adults. At the same time, in accepting the evidence based on a lower level of evidence (than an RCT), the G-BA also acknowledged the special features and limitations in conducting pediatric studies, including the very small number of patients. After considering other factors such as the lack of availability of alternative therapies, the severity and rarity of the disease, the G-BA saw an indication of a non-quantifiable additional benefit within the framework of an evidence transfer.

The AMNOG balance sheet for paediatric medicines

Given the legal and methodological requirements for the benefit assessment of paediatric medicines outlined above, the question now arises as to how such medicines perform in the AMNOG balance sheet and how the enshrined option for evidence transfer is utilized. For this purpose, all completed procedures were recorded in the vfa's AMNOG procedure database. As of November 24, 2023, the database contained 82 completed procedures for pediatric indication extensions (PIPs) and 6 procedures for PUMA approvals, relating to a total of 54 medicinal products. The most common therapeutic areas included endocrine, nutritional, and metabolic diseases.
(32 procedures), infectious diseases (27), diseases of the nervous system (10), and oncological diseases and neoplasms (6). According to the Federal Joint Committee (G-BA), the statutory health insurance (SHI) target population for the PUMA procedures ranged from 980 to an estimated maximum of 86,000 patients.
For PIP indication extensions, the range was between 1 and 20,200 patients with a median of 125. A high proportion of these procedures therefore affected only a small target population.
For the interpretation of these and subsequent figures, it should be noted that among the 88 procedures, 25 procedures alone related to the treatment of cystic fibrosis in new PIP indications with the active ingredient ivacaftor or combinations containing ivacaftor. Thirty of the 88 procedures involved drugs for the treatment of rare diseases (orphan drugs). A review of recent years has shown a significant increase, particularly in AMNOG procedures for new pediatric indications (Fig. 1).
The additional benefit balance is shown in Figure 2. Particularly striking is the proportion of PIP procedures with an unproven additional benefit of approximately 58%, as this is higher than the proportion for all other AMNOG procedures (approximately 47%). Also striking is the small proportion of cases with a quantifiable extent (minor, considerable, or major) of only approximately 6% with a proven additional benefit. Taking into account procedures without orphan drug status, the proportion of PIP procedures with an unproven additional benefit is 74%, and the proportion with a quantifiable additional benefit is only 4%. For PUMA approvals, the significance of the additional benefit extent is less significant given the small number of cases (N = 6).
In 6 procedures for 5 PUMA approvals (propranolol, glycopyrronium bromide, hydrocortisone, melatonin, vigabatrin twice daily), a dossier with an RCT was available for 3 procedures. Evidence transfer was reviewed in 2 procedures (propranolol, hydrocortisone) and considered by the G-BA for 1 procedure (propranolol).
In approximately 45% of the PIP indication extensions (36 of 82 procedures), an RCT justifying approval was available for the benefit assessment. In 37 procedures, the possibility of evidence transfer was discussed in the justification for the decision, and in 18 of these, it was accepted by the Federal Joint Committee (G-BA). However, these findings require separate consideration in the following discussion for a meaningful interpretation.

Discussion of AMNOG assessment practice
Conditions for evidence transfer
Following the positive use case in the vandetanib procedure, it subsequently became apparent that the requirements for evidence transfer in the benefit assessment for PUMA drugs represent a significant hurdle. This was evident, for example, in the subsequent procedure for hydrocortisone (Alkindi®), in which no evidence transfer was carried out because no controlled studies (including in adults) were available compared to individual formulations containing hydrocortisone. This was despite the fact that the G-BA acknowledged in its deliberations that drug development for use in children must follow the pediatric investigation plan approved by the PDCO, which is intended to ensure that children are not exposed to avoidable clinical trials, and a direct comparative study was therefore not part of the approval.
(G-BA 2018a). At the same time, the added value of a safe dosage for infants and children was not recognized as an additional benefit.
In the G-BA's decision-making practice, the EMA's findings on the medical rationale for the transferability of such data represent a sufficient rather than a necessary condition for the acceptance of evidence transfer. They can be decisive for PIP indication extensions (e.g., in cystic fibrosis), but are not in all cases (G-BA 2022, G-BA 2023). While the G-BA acknowledges that approval is based on an extrapolation of pharmacokinetic and pharmacodynamic data, recognition of the additional benefit is not possible on this basis alone (G-BA 2021c). However, a necessary condition for evidence transfer with recognition of the additional benefit is a previously demonstrated additional benefit in the adult population. Another necessary criterion for evidence transfer is the requirement of an identical appropriate comparator therapy for adults and for subsequent pediatric extensions of the drug approval (G-BA 2021b). Deviations in the appropriate comparator therapy can result not only from differences in treatment standards and the availability of approved therapies, but also from rapid changes in healthcare practice in the time interval between benefit assessments (G-BA 2018b). This is particularly sensitive given that the AMNOG procedures for pediatric extensions of an indication only begin, on average, around five years after a drug is first placed on the market. In addition, there are further hurdles regarding the presentation of the comparability of disease courses or the usual requirements regarding the suitability of methodological evaluations or the completeness of all patient-relevant endpoints.

Limitations of the results on evidence transfer
Past procedures show that evidence transfer was accepted by the G-BA in 18 procedures, but only in three therapeutic areas (cystic fibrosis, chronic hepatitis C, and atopic dermatitis). However, it should be noted that these are 11 procedures involving the active ingredient ivacaftor or combinations containing ivacaftor, for which evidence transfer (sometimes based on an RCT) was carried out in a comparable constellation for minor expansions of the approved age groups and genotypes. Four further cases concern the treatment of chronic hepatitis C, which is an exception in the AMNOG (German Medical Device Act) even for adults due to dramatic effects and the inclusion of single-arm studies. For two further procedures involving the active ingredient dupilumab for the treatment of atopic dermatitis, RCTs justifying approval were available for the assessment in the respective pediatric age group (G-BA 2021e, G-BA 2023). In principle, these RCTs could have led to an additional benefit even without evidence transfer, but were only considered as a supplement by the G-BA, particularly because they fell short of the standard requirement for study duration for chronic diseases in the AMNOG of at least 24 weeks. In one of the procedures, the evidence transfer was also limited to a patient group whose clinical picture was sufficiently similar to that of adults (G-BA 2023). Outside of these cases, only one procedure (already described above) on vandetanib in 2017 involved evidence transfer based on non-randomized evidence for children (G-BA 2017).
A limitation of the analysis is that no procedures were systematically recorded for which, due to the respective disease, only children and adolescents (e.g., juvenile psoriatic arthritis) or, in addition to adults, a pediatric subpopulation under 18 years of age (e.g., Hodgkin's lymphoma, melanoma) were included in the approval. Although the latter are more common, they do not represent a pediatric extension of indication in the narrower sense. At the same time, such procedures significantly complicate an analysis of the treatment of pediatric patients, as they are rarely identified as a separate issue by the Federal Joint Committee (G-BA) in the decision (e.g., for approvals of patients 12 years and older). In some cases, however, evidence transfer may also be relevant for such procedures (G-BA 2021a). A further limitation may arise from the anticipated G-BA requirements regarding evidence and evidence transfer, since in some cases pharmaceutical companies do not even seek to demonstrate additional benefit for pediatric indications by referring to the formal claims.

Further requirements for the additional benefit
The specific features and limitations of conducting pediatric clinical trials for the recognition of evidence below an RCT are only mentioned and considered in a few exceptional cases in the G-BA's supporting reasons (G-BA 2015, G-BA 2017, G-BA 2022). Nevertheless, the AMNOG also stipulates the general requirement to submit studies of the highest level of evidence (RCTs) for pediatric approvals. To derive proof of added benefit as the highest possible level of certainty, two RCTs are usually required (as is the case with adults). This requirement should be questioned, especially for pediatric populations.
In addition to the high proportion of unproven added benefits, the analysis also highlighted the very low proportion of quantifiable added benefits for new pediatric indications. Regarding the quantifiable added benefit, all six procedures (as expected) had RCTs justifying approval. For all other PIP indication extensions with an added benefit, the G-BA classified this as non-quantifiable. The reasons for the non-quantifiable extent were the evidence transfer (with and without supportive RCTs), the existing uncertainties of an RCT (e.g., due to small patient numbers), or the orphan drug status of a medicinal product with non-randomized approval data (G-BA 2019, G-BA 2021d). In this context, the "guardrails" enshrined in the Statutory Health Insurance Financial Stabilization Act (GKV-FinStG) since the end of 2022 and their impact on reimbursement negotiations should also be considered. The proportion of procedures with at least a significant added benefit was only 5%. This raises the question of whether not only the requirements regarding the added benefit itself, but also the quantifiability, meet the specific characteristics of studies in pediatric settings.
Another requirement for the proof of added benefit is, as is well known, the presence of therapeutic advantages in so-called patient-relevant endpoints. In this regard, there is no evidence that the specific characteristics of conducting pediatric studies are systematically taken into account in the AMNOG procedures. The only exception to the rule is the consideration of the endpoints "lung clearance index" and "body mass index" in cystic fibrosis (G-BA 2022).

Availability of children's medicines
With regard to availability, it should be noted that of the PUMA drugs examined, vigabatrin (Kigabeq®) was out of distribution at the time of the analysis. The drugs with a PIP indication extension included in the present analysis were available in Germany. However, it should be noted that the AMNOG procedures for pediatric extensions of a therapeutic indication commence several years after the initial market launch of the new drugs. The affected products were therefore generally available in Germany with a reimbursement amount at that time. At the same time, the absolute size of the new pediatric target populations and their share of the total SHI target population of a drug relevant for negotiations were small in many cases. Thanks to this starting point, a predominantly unproven additional benefit for the pediatric indications has not yet led to further market withdrawals in the subsequent negotiation framework. However, it is questionable whether this should be a benchmark for assessing the framework conditions. Rather, the results suggest that the pediatric indication extensions remain available in healthcare despite the rigid framework of the AMNOG benefit assessment. This actually undermines the underlying promotion of the development and availability of pediatric medicines.
outlook
It should be noted that the issue of benefit assessment of pediatric indication extensions has now become a largely ignored norm. However, the relevance of pediatric medicines has recently become more of a focus in German pharmaceutical policy. Following the acute supply problems caused by supply bottlenecks, the Act to Combat Drug Shortages and Improve Supply (ALBVVG) introduced measures to improve the supply of generic pediatric medicines. However, the development and availability of patented pediatric medicines, as well as the challenges of assessment, have not yet played a prominent role in the discussions surrounding the AMNOG.

Appropriateness of the assessments
In its primary function, the AMNOG (Medical Act on Drugs and Healthcare) is intended to contribute to reducing pharmaceutical expenditures as an instrument of price regulation. This now generates savings of several billion euros annually for statutory health insurance funds. However, at least for pediatric indication expansions with a very small target population, the question arises whether conducting a comprehensive AMNOG procedure might not result in more costs for the insured community than the potentially achievable savings. The appropriateness of many such benefit assessments for achieving savings in statutory health insurance appears debatable from a health economics perspective.

Requirements for evidence transfer
In recent years, the G-BA has only applied the evidence transfer tool in a few specific cases and in just three therapeutic areas. The legislator's intention in requiring evidence transfer under Section 5 (5a) of the Ordinance on the Benefits of Pediatric Medicines was to better consider the specific characteristics of pediatric medicines. However, the regulation primarily covers the special case of evidence transfer between patient groups or partial indications. The G-BA's case law applies more to the transfer of evidence from an adult population to an expanded PIP indication. The requirement of a previously proven additional benefit for the older target population appears understandable at first glance. Nevertheless, for the target population of children, the healthcare relevance, the additional benefit, and the underlying endpoints may have independent significance. The requirement of an identical appropriate comparator therapy also needs to be questioned. Both known differences in the availability of approved treatment options for children and the several years between the initial market launch of such procedures will likely lead to more frequent deviations. Overall, the rather rigid requirements for the possibility of evidence transfer should be reviewed.

Consideration of the specifics of therapy situations
Of central importance for promoting the development and availability of pediatric medicines is the consideration of the special features and limitations of conducting pediatric clinical trials during the approval process. These special features have since also been acknowledged in the AMNOG benefit assessment when applying evidence transfer. In recent assessment practice, such references are rarely found in the supporting reasons for G-BA resolutions. While the central objectives of approval and an HTA-oriented additional benefit assessment vary, it is nevertheless worth questioning whether the special features of conducting pediatric trials should be systematically considered as an evaluation criterion within the AMNOG framework. This also applies in the context of the other general requirements in the AM-NutzenV. Accordingly, there are cases in which it is "impossible or inappropriate to conduct or require studies of the highest level of evidence." In these special therapeutic situations, "evidence of the best available level of evidence must be submitted." This requirement has no practical consequences, as the feasibility and appropriateness of clinical trials with the highest level of evidence are not systematically examined, and non-randomized trials are generally deemed unusable (vfa 2023). This applies not only to pediatric medicines under the AMNOG, but also to adults. Especially with pediatric approvals, however, the question arises as to whether the available evidence does not already generally take into account the specific features of conducting pediatric clinical trials (including the avoidance of unnecessary trials) or whether the highest possible level of evidence is already present. In such special cases, the requirements for study design and reliability of results should also be subject to critical discussion.
from:
Dr. PH Andrej Rasch (1)

1: Association of Research-Based Pharmaceutical Companies (vfa)