Pre-proof How should we evaluate the cost-effectiveness of CAR T-cell therapies ?

This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Recent approvals of pioneering chimeric antigen receptors (CAR) T-cell therapies, axicabtagene ciloleucel (Yescarta®) and tisagenlecleucel (Kymriah®), have resulted in considerable interest in the budgetary impact, clinical effectiveness and cost-effectiveness of these technologies. Traditional approaches of appraising new technologies in the UK would mean several promising highly specialised technologies such as CAR T-cell therapies would not gain approval. This article provides insights into the opportunity costs and challenges associated with CAR T-cell therapies, and describes the reimbursement models developed to ensure market access for CAR T-cell therapies.
In 2017, NICE's evaluation guide to the Process and Methods of Technology Appraisal (TA) proposed a threshold of £100,000-£300,000 per QALY for the evaluation of highly specialised technologies (HSTs)(1). The proposal was accompanied with a budget of £20 million per year (2), for each technology evaluated under the technology appraisal (TA) and highly specialised technologies (HST) guidance. The budget is set to signal the need for commercial discussions between the company and NHS England to keep prices low.
Axicabtagene ciloleucel was the first CAR T-cell therapy to be approved for adults living with certain types of non-Hodgkins lymphoma (3). In August 2018, NICE recommended against its use (4) on the basis that the cost-effectiveness estimates of Yescarta compared with salvage chemotherapy exceeded £100,000 per QALY gained, with a high degree of uncertainty in the estimates. Within If future cell and gene therapies have incremental cost-effectiveness ratios beyond the conventional £20,000-£30,000per QALY, what is the drawback? The real concern here is that the cost of these therapies is unsustainable and presents an opportunity cost dilemma to the NHS. For example, the opportunity cost for 200 patients treated with a single dose of CAR T-cell therapy is likely to be up to £56m, which would for example otherwise enable us to treat around 630 patients for lung cancer, the most common cause of cancer death, accounting for around a fifth (21%) of all cancer deaths (7) ( Table 1). Similarly, the opportunity cost to treat 200 patients with a single dose of CAR T-cell therapy could otherwise be spent on 4,435 patients undergoing coronary artery bypass grafting (CABG) to treat coronary heart disease (CHD), the most common type of heart and circulatory disease, and one of England's leading causes of death and single biggest cause of premature death (8). To address challenges of affordability, flexible reimbursement models have been introduced in England to reduce the risk of therapies hitting the market and limiting patient access. These are often embedded with a simple discount, but may comprise more complex models such as performance based models based upon pre-specified clinical outcomes being met (11,12) and annuity models. Performance based models, also known as an outcomes based models, facilitate patient access to new therapies, enabling payers to manage their overall budgets and limit the financial losses as a consequence of treatment failure (13). This model has many variations, including a money back guarantee, a confidential discount and a managed access agreement (MAA). The MAA is an agreement between the payer and manufacturer, specifying the type of data to be collected while treating patients. The data includes, but is not limited to quality-based measures, clinical outcomes, and patient satisfaction (6, 14). Annuity models allow the cost of therapies to be spread over a specified period, allowing the potential to reward innovation and to better align costs with the time period over which benefits are delivered to the patient. The aim is to reduce the up-front budget impact to the payer and reduce the initial cost of treatment (15). Implementation of performance based models with MAA have been linked to tisagenlecleucel and axicabtagene ciloleucel in England, though the exact details are not accessible due to the confidentiality issues.
While, the aim of these models is to facilitate management of budgets and limit the impact of treatment failure, both models require data collection and monitoring over time, which can impose a burden on health services.
Similarly, how other countries in Europe will manage their budgets with the high-cost of CAR T-cell therapies is unclear. Germany have agreed to a performance based deal (16) but we do not currently know how Germany will assess the CAR T-cell therapies as the healthcare system traditionally waivers the MAA (17). Italy, on the other hand is not new to performance based models, particularly for oncology drugs, and has unveiled a model that involves payment in instalments based on patient response to treatment (18). Payments will be made at the time of infusion, after six months, and after 12 months. If the treatment is unsuccessful at any point in the 12-month period, hospitals will not have to make any subsequent payments (19). Additionally, in France Kymriah and Yescarta have been available for prescription through the early access program known as the l'Autorisation Temporaire d'Utilisation (ATU) and is being used for gathering real-world data that will be used during pricing negotiations (17). Finally, while most commercial insurers in the U.S. do cover CAR Tcell therapies, they do so on an individual basis, writing single-patient agreements each time. The staggering prices have led insurers to exercise coverage at a decreased rate, due to high-prices and the cost of aftercare (20).
Challenges around the clinical effectiveness of cell and gene therapies are that data are typically captured over shorter time periods (21), with a small number of patients and hampered by poorly understood study endpoints (22). Moreover, there are difficulties in identifying the appropriate comparator to measure these therapies against (23) and there is limited experience of safety evidence that might occur years after the treatment. These uncertainties make it difficult to produce estimates of health and economic impact that are the core of any assessment of value (24).
In brief, the increase in spending on cell and gene therapies poses an ongoing risk to the NHS that may not be sustainable with larger patient populations. To ensure all potential risks have been captured, NHS England will need to work closely with manufacturers to address concerns around evidence generation and pricing. However, the introduction of new reimbursement models mean NHS England has greater flexibility to address the affordability, though reimbursement models will