Mismanaged Access?

Managed access as a response to regulatory acceleration

For many years, we have seen a trend towards accelerated pharmaceutical regulation based on less mature evidence. Conditional marketing authorisations, adaptive pathways, single-arm trials and surrogate endpoints are now standard features of the oncology and advanced therapy landscape, not exceptions.

This creates a structural problem. What may be sufficient for FDA marketing authorisation, and increasingly so within the UK and EU, is now diverging rapidly from what HTA bodies have traditionally required for reimbursement. The JCA framework, whose methodology is sceptical of precisely the evidence types that accelerated licensing generates, risks compounding the problem at a European level.

The logical policy response is more dynamic HTA evaluation: managed access agreements, coverage with evidence development, outcomes-based arrangements that allow earlier patient access while managing uncertainty through real-world evidence collection and subsequent review. The Cancer Drugs Fund (CDF) and Innovative Medicines Fund (IMF) in England represent one version of this model. France's accès précoce (AAP) system, which provides fully reimbursed early access pending a final HTA determination, represents another.

So far, so familiar: this debate has been running since at least the mid 2010s. But there appears to be a substantial divergence in the experience of managed access from one side of the Channel to the other:  in France, AAP applications are at their highest level since the launch of the scheme; whereas in England, only a single medicine in one therapeutic indication has entered the CDF or IMF over the last 18 months. What are the reasons for this and – more importantly – what lessons should policy makers draw to support sustainable models of access moving forward?

A tale of two systems

The AAP was launched on 1 July 2021, replacing France's previous ATU system. In its first four years, 223 medicine indications were assessed, with an acceptance rate approaching 70%, delivering access to over 140,000 patients who would otherwise have had no available treatment. Overall, 70% of initial applications have received a favourable opinion, and products have remained in the scheme for an average of 17 months before transitioning to routine reimbursement. Industry demand has continued to grow: pre-submission meetings between companies, HAS and ANSM have exceeded 170 since launch, and application volumes have increased continuously year-on-year.

The experience in England looks strikingly different. The current CDF was established in 2016 to expand access to cancer medicines pending the resolution of clinical uncertainty; the IMF, its structural equivalent for non-cancer innovation, followed in 2021. Initially, there was high demand: NICE saw eleven entrants to the Cancer Drugs Fund in 2019/20, and eight in 2021/22. But latterly there has been a dramatic decline: entrants to managed access dropped to two in 2022/23 and zero in 2025/26. Expenditure against the £680m budget across both managed access schemes had declined to around just £175 million in 2024/25.

The contrast is stark: in France, a scheme that did not exist five years ago is receiving record numbers of applications; in England, a scheme that has operated for ten years has effectively ceased to function as an entry route for new medicines.

Loss aversion as commercial leverage

From a company perspective, the commercial logic of the AAP rests on two foundations. The first is temporal: companies set their own price during the early access period, capturing revenue from the point of entry rather than waiting the 500-plus days that the standard French pricing and reimbursement pathway typically requires. Critically, this revenue can be generated at a very early stage: 42% of AAP approvals were for pre-licence indications. The second is structural: the health system's own loss aversion provides the company with meaningful negotiating leverage when the final price determination arrives: the manufacturer arrives with an established patient population and clinical momentum; the payer arrives knowing that refusal carries political and clinical consequences that a clean rejection at the point of licensing never would. The modest 16% average reduction from manufacturer-set AAP price to final reimbursement price likely reflects this dynamic at work.

The model in England offers fewer of these advantages at the entry stage. Pre-licence reimbursement is explicitly ruled out; the requirement for a full entry appraisal erodes the speed of access advantage seen with the AAP; and the requirement for an entry price which NICE would consider plausibly cost-effective also erodes the company’s ability to set a comparatively high price anchor for its exit negotiation. The English health system is therefore better equipped to manage the loss aversion risks, but at the expense of having significantly less attractive terms of entry.

The wrong kind of uncertainty

Where the UK and French managed access systems have a core similarity is in the insistence upon a full HTA appraisal at exit: not just looking at the clinical performance of the medicine, but taking full account of any methodological or comparator changes in the intervening period. This creates a critical policy issue, arising from the fact that not all uncertainty is equal.

Performance uncertainty - will the medicine work in the real world as it appeared to in trials? - is, in Frank Knight's terms, genuine risk. It is uncertain, but knowable in advance. Managed access works in these cases precisely because this type of uncertainty can be defined with enough precision for companies to take the risk of entering. If a defined proportion of patients fail to respond within a set period, the price falls by a predictable amount. The company and the payer are betting on the same, clearly specified game.

Methodological uncertainty - will the medicine be assessed against the same criteria and ruleset? - is categorically different. It is closer to Knightian uncertainty proper: neither side can specify in advance what might change, because the change is a political or institutional decision outside the company's control rather than a clinical outcome being actively measured. Nobody can price a bet whose rules might be rewritten after the wager is placed, and that unpriceable risk tends to freeze commercial decisions rather than merely discount them. This methodological uncertainty is compounded by the asymmetric risk of comparator price decay during the managed access period as established products lose exclusivity, with a consequential downward price pressure even if the medicine under assessment has achieved its expected clinical outcomes.

However, where the UK and French systems diverge is in the actual realisation of these uncertainties. In England, the 2022 NICE methods update materially altered the valuation of medicines already in managed access, particularly as a result of the replacement of the ‘end of life modifier’ (which provided a value uplift focussed upon cancer medicines) with a ‘severity modifier’ (which applied more evenly across cancer and non-cancer indications). The result was medicines exiting managed access with ICERs that were up to 70% higher than previously assumed – even in circumstances where their clinical endpoints had been met. The realisation of methodological uncertainty trumped the successful management of performance uncertainty. By contrast, no mid-flight shift in HAS methodology during the AAP period has yet occurred in France at the scale experienced in England. The lived experience of French affiliates therefore aligns with that of England pre-2022: methodological uncertainty is a theoretical risk which is heavily discounted, and more than exceeded by the strong commercial incentives to participate in the AAP process.

The result of these divergent experiences can be seen from the numbers: the drop off in managed access entry in the UK corresponded with the change in NICE methods in 2022; demand for the AAP continues to grow in France. But it is more than likely that if and when a comparable shift occurs in France - whether driven by fiscal pressure, JCA alignment, or a reassessment of ASMR methodology - the chilling effect on AAP applications will be similarly rapid and significant.

A pragmatic way forward?

The UK Government has recently announced a pilot to change the entry process into managed access, with the option for direct entry into the CDF or IMF without an entry appraisal. Yet if, as expected, this new process requires manufacturers to sign up to a commercial ‘true-up’ to the cost effective price determined by the exit appraisal, we are unlikely to see an improvement to managed access entry rates. The problem is that a ‘true-up’ compounds the existing price uncertainty: it means company revenue during the managed access period is now at risk, as well as future revenues on exit.

In reality the change needed is conceptually straightforward, even if economically impure. The entry decision should establish the terms for the exit HTA appraisal: the comparator, the endpoints that will be used at exit, the evidence standard required, the threshold applicable. The exit process should then be scoped exclusively to resolving the specific clinical uncertainties identified at the point of entry, with other parameters held constant. Doing so would enable manufacturers to take advantage of the huge benefits that ongoing real world data collection provides, without the fear that the appraisal goalposts will move whilst they are doing so.

There is plenty to critique in this approach. Many will argue that it is unreasonable for two products in the same therapeutic area to be assessed on a different basis at the same time, simply by virtue of one exiting managed access and another entering directly into standard reimbursement. This requires a conceptual shift by HTA bodies and payers from considering managed access as an ‘interim’ reimbursement decision and towards it being a ‘standard’ reimbursement decision with conditionality. Given around 90% of products successfully transition to routine reimbursement in England, this shift does not appear to be a significant intellectual stretch.

Others will argue that this change doesn’t address the main problem: the high level of evidential uncertainty at the point of entry. But here, commercial tools can support resolution: performance-based payment models or risk shares based on the final price determination have both been deployed in precisely these circumstances. Crucially, though, such tools struggle to manage both clinical performance risk and methodological uncertainty. This highlights the criticality of fixing the methodological part of the equation so that clinical uncertainty can be managed commercially on a more routine basis.

Finding the sweet spot

For companies navigating global launch sequencing, the use of managed access will be an increasingly important tool. The experiences of the UK and France offer fascinating observations: that companies have low tolerance for value uncertainty, that methodological or comparator changes create the greatest uncertainties, and the actual experience of this uncertainty becoming manifest will have a dramatic chilling effect on the attractiveness of managed access approaches. The policy implications are important: at a point in time when regulatory acceleration requires greater HTA dynamism, there is a strong case to fix the parameters for the exit evaluation upfront in order to defend or rebuild trust that value is being appropriately shared.

Robert Kettell is a Partner at Newmarket Strategy and previously served as Director of Medicines Negotiation and Managed Access at NHS England. This article reflects his personal views.

‍