For the next blog post in the series where I review papers published as part of the “Economics of Gene and Cell Therapies” in PharmacoEconomics, I review: “Innovative Payment Models for Sickle-Cell Disease Gene Therapies in Medicaid: Leveraging Real-World Data and Insights from CMMI’s Gene Therapy Access Model” found at the following link: https://link.springer.com/article/10.1007/s40273-025-01474-3. 

The authors developed a cost-effectiveness model using sickle-cell disease as a case study to assess the financial impact of the following payment methods: outcome-based agreements (payment linked to patient outcomes), volume-based rebates (discounts on quantity purchased), and guaranteed rebates (refunds based on pre-determined performance metrics). These three models were proposed by the Center for Medicare and Medicaid Innovation as potential innovative contracts for gene therapies and the authors wished to explore the impact of the choice of contract. 

The authors used Colorado Department of Health Care Policy & Financing (HCPF) 2018-2023 data in a three-state (sickle cell disease, stable and dead) Markov model to estimate the financial impact on payers of the three payment methods. Current standard of care was compared with exagamglogene autotemcel (exa-cel) and lovotibeglogene autotemcel (lovo-cel). Modelled patient characteristics aligned with those used in sickle cell disease clinical trials with adjustments made to the modelled population to ensure that the modelled cohort was eligible for gene therapy which the authors defined as experiencing a set number of vaso-occlusive events. 

Scenarios were conducted to assess the impact of varying eligibility criteria for innovative payment models which included assessing uncertainty around treatment effectiveness and durability. All costs were taken from the Colorado database; eligible patients were identified using ICD-10 codes. The study analysed data over the course of 6 years to assess the impact of the three innovative contracts. The authors present the following findings:

• Outcome-based agreements: these may offer payers financial savings although it varies based on response rates with higher savings with lower response rates. An extension of a contract from the standard 6 years increases the likelihood of a payback.
• Volume-based rebates: The authors varied the eligibility for gene therapy to assess the impact of sales volume on payer budget impact. Increased volume leads to higher rebates. 
• Guaranteed rebates: Higher guaranteed rebates generate higher savings for payers. 

There is limited information provided on the cost-effectiveness model but based on what is described, the authors simulated patients with sickle cell disease being treated with gene therapies compared to standard of care. The model was a basic model consisting of two health states and an absorbing state: sickle cell disease, stable and dead. A comparison between this and published models indicates that this seems to be an over-simplification of the sickle cell disease pathway, as acknowledged by the authors (1,2). The modelled patient profiles were based on patients participating in clinical trials suggesting that they may not be representative of the real-world. However, costs were based on real-world all-cause Colorado claims data. The described model lacks transparency as it is not clear how transitions between health states were calculated. 

This modelling approach stated by the author’s as “chosen to maintain focus on the study’s key objectives within the available time frame and data” may limit the usefulness of the findings. However, a key question is why a cost-effectiveness analysis was needed to answer the study’s key objectives. Ultimately, the authors are seeking to understand budgetary impacts meaning that a simple budget impact assessment would have been more suitable than an (over-simplified) cost-effectiveness model. The authors could have defined a population eligible for gene therapy – accounting for subgroups if needed – based on the Colorado claims data and thus modelled a real-world population with costs also derived from the database in a budget impact assessment. As cost-effectiveness models are developed to assess the value for money of an intervention rather than budgetary impact, it is not clear why this approach was taken. Indeed, the authors refer to budget impact in the key points section of the article which raises questions around the approach. 

The results of this US-based study are intuitive:

• The higher the number of patients who are eligible for gene therapy, the higher the volume-based rebate; 
• The higher the rebate agreed, the higher the savings and; 
• Outcome-based agreements should be viewed as a long-term agreement to be of value to the payer. 

The findings are also likely transferable to other regions despite being derived from US-based insurance claims data; however, it is clear that both developers and payers prefer simple payment models. Further, these findings confirm a known issue with outcomes-based agreements whereby a relatively long-time frame is required for them to be useful for payers. Combined with their resource intensive nature, this makes them an option only when every other innovative payment option is not feasible. 

In conclusion, this study used a sickle-cell disease case study to assess the financial impact of outcome-based agreements, volume-based rebates, and guaranteed rebates. The methods are unclear, but the authors confirm what is already known by payers: an outcomes-based agreement is a last resort requiring long-term follow-up. 

1. Lopez et al., 2025. Cost-effectiveness of exagamglogene autotemcel gene-edited therapy in patients with sickle cell disease with recurrent vaso-occlusive crises in the United States. J Med Economics; 29: 547-62.

2. Dias et al., 2026. Cost-Effectiveness of Hydroxyurea for Treatment of Children With Sickle Cell Anemia in Ghana. Value in Health; 55: 101586