Abstract
Introduction
Newly marketed pharmaceuticals command increasingly high prices, particularly in oncology where single treatment cycles can be very costly, even though data at market approval are usually inadequate to describe thoroughly their effectiveness and cost-effectiveness. 1 Both manufacturers and payers have shown interest in solutions to achieve rapid access for patients – known as ‘market-access agreements’. Although the literature offers a variety of definitions,2–5 these agreements can be roughly grouped under two main types: 6 financial-based and performance-based. Financial-based schemes focus on the budget impact of a new drug and usually consist of price/dose discounts. Performance-based schemes involve collecting clinical evidence once a new drug has been marketed in a healthcare system and lead to payment only for patients responding to therapy, an objective which is methodologically and practically challenging while not without political appeal.
The underlying strategy is to help reduce the likelihood of payers adopting technologies that turn out not to be cost-effective, while at the same time helping manufacturers increase market access time and take profit. 7 Market-access agreements can also be considered an opportunistic way for manufacturers to keep official prices high such as maintaining a common European price, 6 while granting health authorities confidential rebates, thus preventing parallel trade and undermining effective external reference pricing.
The English and Italian health services have been the first countries to introduce market-access agreements on specific drugs. These contracts are now called Managed Entry Agreements (MEA) in Italy and Patient Access Schemes (PAS) in England and Wales. We focus on the area of oncology, where many new drugs are costly and fall under these arrangements. 8
Italy
Framework
AIFA, the Italian medicines agency, signed its first contract in July 2006, 9 and now 44 are in force for 33 drugs, 10 37 of them in oncology. AIFA can stipulate three different types of agreement with pharmaceutical companies during price and reimbursement negotiations for new drugs or indications. ‘Cost-sharing’ contracts involve a price discount, usually limited to the first month or cycles of therapy; these discounts are normally monetary and manufacturers are expected to make a rebate payment. The other two contracts are performance-based: for each non-responding patient, the manufacturer is expected to rebate the full price for ‘payment-by-result’ or part of it for ‘risk-sharing’ contracts. If a patient meets ‘non-responder’ criteria, the hospital pharmacist has to apply to the manufacturer for pay-back not later than the end of the year; then the manufacturer can accept or reject the request (requiring arbitration).
The complex management of these schemes is entirely based on web registries owned by AIFA. 11 Hospital consultants are required to complete on-line prescription forms, with the patient’s identification data, therapeutic indication and dosages. In addition to the prescription, the registry requires the treating physician to record follow-up clinical data and outcomes.The system validates each prescription and automatically requests the hospital pharmacy to release the drug.
Assessment
In September 2013, AIFA published the revenues of MEAs for the first time. 12 The total expected pay-back was €46.3 million in 2012. However, one-third was subject to disputes with pharmaceutical companies (22%) or late requests by hospitals (11%). €31.3 million were eventually collected, around 5% of the total expenditure for the indications of the drugs under MEA. 7
Management costs for MEAs are also substantial. In May 2012, AIFA awarded a three-year tender for €8.7 million to a leading private international consulting company to manage the registries. In practice, this led to an interruption in service at the end of 2012, because of incompatibility between the old and the new information technology systems, which continued in 2013 and may undermine most of the rebate for that year. Furthermore, AIFA reports on its website that around €1 million should be added to the total direct cost. 7 Finally, the NHS hospital consultants’ and pharmacists’ time for completing the forms also represents a cost. AIFA has published on its website a very precise estimate of the time required by health professionals to complete the most important forms. However, the negligible average times suggested (less than 1 min) do not appear to be realistic. 13
Despite AIFA expectations for the MEAs’ contribution to the clinical evidence base, no published report has yet included data on drugs subject to performance-based agreements. Going through the forms referring to the patients’ clinical status, 10 it is clear that they closely reflect the approval indication, requiring very little additional information and thus not useful for an extended clinical assessment. 7
England and Wales
Framework
The English Department of Health (DH) agreed its first PAS with a pharmaceutical company in October 2007. Currently, the NHS has signed 42 contracts covering 32 drugs, 14 almost half of them (19) anti-cancer therapies.
The 2009 Pharmaceutical Pricing Regulation Scheme (PPRS) – the voluntary agreement negotiated periodically between the DH and the Association of the British Pharmaceutical Industry (ABPI) – was the first to recognise PAS as a scheme to be used exceptionally for improving the cost-effectiveness of innovative drugs and making their cumulative burden on the NHS manageable. 15 A pharmaceutical company can propose any kind of PAS to the DH, which will take a decision with input from the National Institute for Health and Care Excellence (NICE), the public agency that assesses new health technologies on behalf of the NHS. The NICE’s Patient Access Scheme Liaison Unit (PASLU) advises the DH whether the proposed PAS is feasible within the NHS in England and Wales. Manufacturers can decide to propose a PAS after market approval or submit it as a response to a negative draft/final guidance by NICE.
Only one type of performance-based contract exists, the ‘response scheme’ (pay-back on drug costs based on response to treatment measured by a test), while five different options are in force for financial-based contracts: 14 (i) ‘simple discount’ (discounted official price of the drug), (ii) ‘free stock’ (first cycles/weeks provided free of charge or higher dosages provided for the same price as lower dosages), (iii) ‘dose cap’ (drug provided free of charge after a specific treatment period), (iv) ‘rebate’ (rebate of a proportion of the per-patient drug dose), (v) ‘single fixed price’ (lump sum irrespective of the duration of treatment).
Hospital pharmacies usually manage PASs without any specific financial support from the DH, tracking each patient’s cycles or clinician’s responses and then computing the rebated price paid by the NHS. The procedures have to be completed within deadlines that may vary substantially between contracts.
Assessment
PAS are commercial-in-confidence contracts whose financial results are not made public, so unfortunately there is little official information. 16 We found only two small, not very recent surveys involving hospital pharmacists that tried to assess the NHS administrative burden of PAS in England and Wales.17,18 Both studies concluded that, in order to prevent missed claims, the NHS did not have the capacity to manage more PAS without funding staff time to manage, co-ordinate and track the contracts. In general, schemes linked to measurement of a clinical response took longer to administer than the pure financial ones (e.g. 45 min per patient for cetuximab and 38 min for bortezomib vs. 18 min for erlotinib and 19 min for sunitinib) 17 and tended to raise more problems. Although hospital pharmacists did not indicate any specific preferred scheme, simpler contracts with fewer requirements for data collection and monitoring were recommended, possibly based on easy financial discounts. 18
Although the 2014 renegotiated PPRS 19 has confirmed that PAS should be the exception rather than the rule, almost one-quarter of all positive guidance issued by NICE included a PAS.
Comparative analysis
Table 1 summarises the number and types of MEAs and PASs on anti-cancer drugs retrievable from the websites of AIFA
10
and NICE
14
and approved up to June 2014. Figure 1 shows the trend over time by type of MEA and PAS in oncology. All PASs introduced in England and Wales after October 2011 have been based on simple discounts. In Italy, even though the financial-based MEAs have proved easier to manage and more efficient than performance-based ones in producing revenues,
7
the latter have by far overtaken the former in the last few years. Table 2 shows the 12 anti-cancer drugs (and their 13 therapeutic indications) under MEA and PAS in the two countries. Only four (azacitidine, erlotinib, nilotinib and sunitinib) come under a common type of scheme (financial-based); the only drug under a performance-based PAS in England and Wales (bortezomib) is managed by a simple financial-based MEA in Italy.
Trends of MEAs and PASs in oncology by type (up to June 2014). Number and types of existing schemes in oncology. MEA: Managed Entry Agreement; PAS: Patient Access Scheme. Anti-cancer drugs under MEA and PAS. CS: cost-sharing; PbR: payment-by-result; SD: simple discount.
Both countries have gained experience with market-access agreements but their practical application has radically diverged over time. AIFA started mainly with financial-based contracts and then shifted to performance-based schemes, while the DH agreed only financial-based schemes in the last few years. The English NHS appears to have opted for cost-minimisation of their management, the simplicity of the schemes prevailing over the possible additional clinical information that might be collected by PASs. The Italian NHS, however, has insisted in investing on performance-based MEAs despite the lack of clinical added value so far.
The major, common concern around these schemes in the two countries is their burdensome administration, mainly borne by the local health professionals involved in compiling patient data and applying for money back from pharmaceutical companies. While the centralised and disruptive management of MEAs has further highlighted their substantial costs, the lack of transparency on commercially confidential PASs means there is no way of taking stock of their experience.
Policy implications
We conducted a comparative analysis of market-access agreements in oncology in the Italian and English NHS according to the existing information, in order to assess their experiences almost a decade after their introduction.
We focused on anti-cancer drugs because there are many contracts in this area. The emotive nature of cancer 20 makes it difficult for health agencies to resist calls for reimbursement of even extremely expensive drugs with marginal efficacy.1,11 Although equivocal results from clinical trials have encouraged policy-makers to put these drugs under performance-based contracts and to collect further evidence and improve their cost-effectiveness, these schemes contribute little to robust clinical assessment in practice, given the absence of randomisation and the uncertain relationship between short-term surrogate and hard end-points and their timings. Simple financial-based contracts seem more efficient as a means for health services to reduce outlay on costly anti-cancer drugs and achieve access for patients.
We may reflect who gains most from market-access agreements. The manufacturer achieves access for a drug priced at a level that maximises the profits which may be taken from a health system and is kept commercial-in-confidence. The health system makes available to its patients costly but modestly effective drugs without paying their full list price. Who gains most depends very much on how these contracts are negotiated, designed and managed. However, the losers of the situation are likely to be small and low-income countries that do not have well-organised procedures or sufficient bargaining power to negotiate lower prices. When market-access agreements lead to prices which are not transparent, these schemes simply penalise those countries which rely on external reference pricing. This lack of transparency raises an issue of public interest on the international level.