Price and reimbursement of advanced therapeutic medicinal products in Europe: are assessment and appraisal diverging from expert recommendations?

Background Advanced therapy medicinal products (ATMPs) represent an important cornerstone for innovation in healthcare. However, uncertainty on the value, the high average cost per patient and their one-shot nature has raised a debate on their assessment and appraisal process for pricing and reimbursement (P&R) purposes. This debate led experts providing for recommendations on this topic. Our primary objective is to investigate the ATMPs P&R process in the main five European countries and to understand if this process is consistent with published P&R expert recommendations. We also investigated the current ATMP pipelines to understand if future ATMPs will create challenges for their P&R process. Methods P&R framework for ATMPs in the European Major five (EU5) countries was investigated through a literature search on PubMed, institutional websites of National Health Authorities and grey literature. The ATMPs pipeline database was populated from a clinical trial database (clinicaltrials.gov), relying on inclusion and exclusion criteria retrieved from the literature. Results Reimbursement status of ATMPs is different across the EU5 countries, with the exception of CAR-Ts which are reimbursed in all countries. Standard P&R process in place for other medicinal products is extended to ATMPs, with the exception of some cases in Germany. List prices, where available, are high and, tend to be aligned across countries. Outcome-based Managed Entry Agreements (MEAs) have been extensively used for ATMPs. Extra-funds for hospitals managing ATMPs were provided only in Germany and, as additional fund per episode, in France. The accreditation process of hospitals for ATMPs management was in most countries managed by the national authorities. As far as ATMPs pipeline is concerned, ATMPs in development are mostly targeting non-rare diseases. Conclusions Expert recommendations for ATMPs P&R were partially applied: the role of outcome-based MEAs has increased and the selection process of the centres authorized to use these treatments has been enhanced; additional funding for ATMPs management to accredited centres has not been completely considered and annuity payment and broader perspective in cost considerations are far from being put in place. These recommendations should be considered for future P&R negotiations to pursue rational resource allocation and deal with budget constraints.

ATMPs are often used to treat severe diseases associated with considerable societal costs [3,4].
Clinical trials are characterized by small patient populations, short investigational duration and single-arm trial design. This is supportive in accelerated procedures granted to medicines for rare diseases with a high unmet need but creates problems for Health Technology Assessment (HTA) [3,5]. ATMPs are one-shot costly therapies, whose benefits can only be appreciated in a longer-term perspective. The temporal misalignment between incremental and avoided costs represents a challenging issue for budget constraints; payers may focus on the shortterm economic impact of medicines, disregarding saving in the long-run [6]. Despite the overall impact on budget being similar to that of a cheaper treatment for a larger patient population, the emotional impact of high prices is stronger and may undermine the principle of equity [4]. In addition, ATMP management and administration is complex and requires: (i) a clear definition of centres of excellence with high-quality equipment and expertise of health-care professionals; (ii) proper funding of the centres, and (iii) facilitated patient access to effective new therapies [6].
These issues have been widely discussed by the literature, driving various recommendations by experts on the assessment and appraisal of ATMPs. In general, experts have not supported a specific value framework for ATMP, but they have recommended to (i) collect more robust evidence [6]; (ii) increase awareness on the value of ATMPs in order to overcome prejudice or excessive unjustified optimism [6]; (iii) strengthen the early dialogue between HTA authorities, payers and other stakeholders, including patients [6]; (iv) conduct post-marketing assessment programmes (which may include registries or observational studies) with the purpose to gain additional information on the efficacy and safety profile of ATMPs [6]; (v) rely on outcomebased Managed Entry Agreements (MEAs) to manage uncertainty on benefits and on annuity payment to split treatment costs over time [6]; (vi) work on an early definition of the criteria for a proper identification of the centres of excellence [6]; and (vii) to adopt a societal perspective in the economic evaluation of ATMPs (or at least a double reference case approach) and face challenges posed by ATMPs to cost-effectiveness analysis, including discount rates and modelling [3,5].
The primary objective of this paper is to investigate how the P&R of ATMPs has been managed in the European Major five (EU5) countries and whether it is consistent with published expert recommendations.
We also compared the current ATMP pipelines with already marketed treatments with the aim of identifying the issues that may be challenging in the P&R process in the future.

Materials and methods
A literature review on the ATMPs P&R status and regulatory frameworks across EU5 was conducted using PubMed (as scientific literature source), the institutional websites of the National Health Authorities for each of the five EU countries [7][8][9][10][11][12][13][14], and grey literature (google and non-peer review journals) using as cut-off date July 2020. ATMPs reimbursement status was defined as "reimbursed ATMP", "not reimbursed ATMP" or "ongoing evaluation" and was assessed searching the institutional websites.
For pipeline analysis interventional clinical trials on ATMPs were retrieved from the clinicaltrials.gov database (www. clini caltr ials. gov) with a temporal limitation from July 2017 to November 2019. Hanna et al. 2016 keywords were used for a preliminary screening of ATMPs trials (i.e. trials focused on ATMPs) [15]. This first screening was validated using Committee for Advanced Therapies (CAT) criteria [16]. Only trials that met inclusion and exclusion criteria described in Table 1 were considered.
Rarity of the targeted diseases in each trial was searched on the Orphanet website [17]. Based on the European Medicine Agency (EMA) definition [18], diseases were considered rare if their prevalence does not exceed 5 cases in 10,000 people. Type, Therapeutic area, Title, URL. Duplicates and trials on investigational products other than ATMPs were excluded and finally the database was analysed according to ATMPs type, trial status, funding origin, therapeutic area and disease rarity ( Table 2).

Results
The following two sections describe the current P&R process for ATMPs in EU5 countries and the pipelines. Consistency of the former with the recommendations of the literature and problems posed by the latter to future P&R negotiations will be discussed afterwards.
The P&R request was not submitted for all ATMPs in each of the EU5 countries: (i) no request for Imlygic submission in France, Spain and Italy; (ii) no submission for Strimvelis in France; (iii) no submission for Spherox in Italy and France (Table 3).
All ATMPs were assessed as medicines and followed the "traditional" appraisal procedure in each country, except for Germany. In Germany, the Federal Joint Committee (G-BA) first categorizes the ATMP either as a medicine or a medical procedure. If the ATMP has pharmacologic properties and its clinical outcome is not dependent on the healthcare professional skills, it is categorized as a medicine and undergoes the benefit assessment procedure according to the AMNOG (Arzneimittelmarktneuordnungsgesetz, Medicine Market Reorganization Act), where discounts on list prices determined by the MAH are negotiated on the grounds of different variables, including the added therapeutic value. Otherwise, it is categorized as a medical procedure and is normally assessed by the PEI (Paul-Ehrlich-Institut) [34][35][36][37]. Spherox and Holoclar were evaluated via the medical procedure; all other ATMPs underwent the AMNOG process (Table 3).
In United Kingdom, the National Institute for Health and Care Excellence (NICE) evaluates ATMPs through     [39][40][41]. Patient access to a new product is possible also before MA with the Authorization for Temporary Use (ATU). To date, Kymriah, Yescarta, and Zolgensma obtained ATU; the ATU for Zolgensma is the only one currently active. During the ATU period, the MAH is allowed to set a launch price before the negotiation with the CEPS.
In Spain, the Agencia Española de Medicamentos y Productos Sanitarios (AEMPS) performs a clinical assessment of new medicinal products and then the General Directorate for Medicines of MoH (Dirección General de Cartera Básica de Servicios del Sistema Nacional de Salud y Farmacia, DGCBSF) prepares a price proposal. Negotiation of the proposal occurs between the Price Committee (Comisión Interministerial de Precios de Medicamentos y Productos Sanitarios, CIPM) and the MAH [42]. The final price is the maximum reimbursable price nationwide, subject to further reduction by hidden discounts negotiated by the Regions [43], due to the fact that the Spanish Healthcare System is decentralized and each Region has the authority to re-negotiate the prices.
In Italy, the assessment and appraisal process is managed by the Italian Medicines Agency (Agenzia Italiana del Farmaco, AIFA), through its Scientific Technical Committee (Commissione Tecnico Scientifica, CTS) and its Price and Reimbursement Committee (Comitato Prezzi e Rimborso, CPR). The CTS evaluates the clinical added value, the place in therapy of the new medicine and possible price comparators. In the case of a positive opinion, the CPR negotiates with the MAH the price and any MEAs [44]. If the product is reimbursed and the price is negotiated at a national level, a request for inclusion in the regional and/or hospital therapeutic lists has to be submitted. The MAH may also ask for the recognition of the "innovative status" of the medicinal product. Innovativeness is appraised on the grounds of three criteria: unmet medical need, added therapeutic value and quality of available evidence and lasts maximum 3 years [45]. Innovative medicines benefit from dedicated funds and immediate access to regional lists, and they are not subject to temporary price reductions by law. Currently, only three ATMPs (Strimvelis, Kymriah, and Yescarta), have been granted this status (for Strimvelis the innovativeness status has expired).
Only Chimeric Antigen Receptor-T cells (CAR-Ts) therapies are reimbursed in all the EU5 while the decisions regarding other products are quite heterogeneous (Table 3).
List prices, on the other hand, are substantially aligned across all the EU5 (Table 3), despite in Italy, Spain and France they are negotiated together with the reimbursements status, whereas in Germany and England prices are freely determined by the relevant MAH, but indirectly regulated (in Germany a discount is negotiated after 1 year of marketing; in England a threshold range for the incremental cost-effectiveness is set, thus addressing pricing strategies by the pharmaceutical companies).
ATMPs were also quite disruptive for MEAs trends in almost all the EU5. Outcome-based agreements, in the form of performance-linked reimbursement, had never been signed centrally in Spain before a "payment at result" for the two CAR-Ts; in Italy, a similar agreement as the one above for the two CAR-Ts was signed, despite the outcome-based agreements being gradually substituted by simple discounts. An instalment plan, with 315,000 € paid up front and four additional annual payments only if the treatment continues to be effective, was applied for Zynteglo in Germany [46]. In England, NICE has created a Cancer Drug Fund (CDF), as a time-limited managed access agreement, which guarantees a faster and temporary reimbursement for cancer medicines, conditioned by evidence-based results which allow a routine appraisal. The two CAR-Ts were included in the CDF, due to the uncertainty of their benefit and costeffectiveness profiles.
Germany and France adopted extra-cost coverage systems to support the management of ATMPs in hospitals. In Germany, when a medicine is used in a hospital setting and its price is not totally covered by an existing diagnosis-related group (DRG) fee, it is possible apply for an extra-budgetary reimbursement, the NUB (Neue Untersuchungs-und Behandlungsmethoden, new examination and treatment method). A status of NUB 1 is granted to a medicine or procedure which is new, innovative, for a low number of patients and which requires higher resource use (other than that covered by the DRG); this means that an extra-coverage can be negotiated by hospitals [47,48]. In France, an additional 15,000€ (fixed) was added to the current DRG fee exclusively for Kymriah and Yescarta ( Table 3).
As far as other polices that affect patients access to ATMPs, in most cases, the accreditation criteria for centre selection are established by the National Authorities, usually through the JACIE (Joint Accreditation Committee ISCT-Europe and EBMT) accreditation process and with the involvement of multidisciplinary medical teams [49][50][51][52][53]. In England, the National Health Service has been working with the JACIE and life sciences companies to get centres up and running. In Spain, for CAR-T, a group of experts for the definition of criteria for designation of centres was established. In Italy, the accreditation process is usually managed by the regional governments, regulated by the minimum criteria for centres' selection drawn up by AIFA for ATMPs.

Pipeline analysis
According to the previously mentioned inclusion and exclusion criteria, 6,982 trials were identified. After the exclusion of duplicates and all studies not targeting ATMPs, 249 clinical trials were included in the final analysis.
The study target was a rare disease in only 114 (46%) trials, while the target disease was not rare in the remaining 135 (54%) (Fig. 3).

Discussion
This paper aimed at comparing assessment, appraisal and P&R process for ATMPs in the main European countries and whether this process is consistent with the expert recommendations currently available.
This study shows similar common patterns. ATMPs follow the same procedure already in place for other medicinal products in all countries, with the exception of Germany. Since the medicines P&R process is different across countries, this makes the reimbursement status of ATMPs heterogenous, with the exception of CAR-T therapies, which have been granted reimbursement in all countries due to the high expectations of their clinical benefits. Furthermore, for some ATMPs, request for P&R was not submitted in all countries. List prices tend to remain aligned in order to avoid crossreference pricing. Our results also indicate that outcome-based MEAs and discounts are extensively used due to the uncertainty on the risk-benefit profile of  ATMPs and their sustainability over time, respectively. This makes net prices divergent from list prices. Comparison of discounts on net prices is not possible as they are not published in any of the EU5 countries, with the exception of Germany. Additional funding for managing ATMPs was provided only in Germany (through the ordinary process of NUB classification) and in France (15,000 € for each DRG) for CAR-Ts alone.
Assessment and appraisal for P&R are partially aligned with the expert recommendations. Payers show a higher availability to draw up outcome-based MEAs, centrally or locally. On the other hand, other expert recommendations have not been applied systematically: ad hoc additional coverage for ATMP administration and patient management was provided only in France, with limited funding. The two delayed payments at results MEAs in Spain and Italy differ from each other both in terms of payment schemes as well as evaluated overall outcome. To date, no country has applied annuity payments, apart from Germany (and one product). We did not find concrete evidence of anticipated early dialogue among stakeholders or a more holistic evaluation of the social impact of these diseases considering the avoided productivity loss.
Pipeline analysis showed that the role of ATMP-based therapy is rapidly evolving from a niche-based setting in rare diseases to a vaster application in diseases which involve larger target populations. In this potential scenario of aiding a larger population, the impact on budget will be higher, if unitary prices remain very high as they are for current ATMPs.

Conclusions
The P&R process for ATMPs in the largest European countries is similar to the one adopted for other medicines. This is aligned to what experts have recommended, i.e. not creating an "ad hoc" framework for P&R of ATMPs, but adapting the assessment and appraisal processes for medicines to ATMPs.
However, some recommendations have not been sufficiently pursued, by domestic payers, including the suggestion of: (i) promoting early dialogue among HTA bodies, payers, the industry, and other stakeholders; (ii) considering, more than for other medicines, the longterm impact and using societal perspective in determining value for money and impact on budget; (iii) introducing annuity payment schemes, which would mean to actually turn the current expenditure into investment.
For the future, since many new ATMPs are going to be launched for non-rare diseases, it is important (i) evaluating in advance the organizational impact of ATMPs and providing health-care centres with the necessary resources; (ii) estimating the budget impact of ATMPs through an appropriate horizon scanning activity; and