Gibson Eddie J, Begum Najida, Koblbauer Ian, Dranitsaris George, Liew Danny, McEwan Phil, Yuan Yong, Juarez-Garcia Ariadna, Tyas David, Pritchard Clive
Wickenstones Ltd, Abingdon, UK.
Augmentium Pharma Consulting Inc., Toronto, ON, Canada.
Clinicoecon Outcomes Res. 2020 May 6;12:241-252. doi: 10.2147/CEOR.S238725. eCollection 2020.
Existing economic model frameworks may not adequately capture the atypical treatment response patterns in immuno-oncology (I-O) compared with conventional therapies and thus may fail to represent the full clinical value associated with disease dynamics and improved survival.
A cost-effectiveness analysis (CEA) of the I-O Regimen (nivolumab/ipilimumab) versus ipilimumab alone in advanced melanoma was carried out by applying a 5-state partitioned survival model (PSM) as a case study, to explore the I-O treatment response and clinical outcomes. The findings were compared with those of a conventional 3-state PSM.
The case study extends the conventional 3-state PSM, by separating the pre-progression state into non-responders and responders, and the post-progression state into normal and I-O progression to account for delayed treatment effects preceding clinical response. Model states were populated using patient-level data (where possible), mapping from the best overall response (BOR), and survival analysis with flexible and traditional parametric methods. Survival functions were applied to progression-free survival (PFS) and overall survival (OS) endpoints across treatment arms using the 4-year follow-up data (data available at the time of the research; since then 5-year follow-up data have been published) from the CheckMate 067 trial. Information on BOR was used as a means of differentiating the I-O treatment response in addition to the outcomes of progression-free and progressed disease. A UK National Health Service and personal social services (NHS/PSS) perspective over a lifetime horizon was used with outcomes discounted at 3.5% annually.
The 5-state PSM generated an increase in quality adjusted life years (QALYs) in both treatment arms and gave a more granular description of patients' health profiles compared with the traditional 3-state PSM. The incremental QALY increased by 13% (from 2.62 to 2.95 QALYs) and the incremental cost decreased by 12% (£29,125 to £25,678) with the 5-state model. In both models, the Regimen had an incremental cost-effectiveness ratio (ICER) relative to ipilimumab alone within the lower bound of the National Institute for Health and Care Excellence (NICE) reference range (£20,000 per QALY gained).
A 5-state economic model, incorporating relevant I-O health states, can be more informative to gain insight into treatment response and progression differences that are not commonly captured in existing economic models. Clinical trial endpoints, including those relating to treatment response, which are not directly reported in ongoing I-O trials, can be mapped on to the proposed modelled health states (although assumptions are required to do so). Improvements in reporting treatment response in future I-O clinical trials could help to further validate and improve the proposed model framework.
与传统疗法相比，现有的经济模型框架可能无法充分捕捉免疫肿瘤学（I-O）中非典型的治疗反应模式，因此可能无法体现与疾病动态和生存改善相关的全部临床价值。
以一项5状态分区生存模型（PSM）为例，对晚期黑色素瘤中I-O方案（纳武利尤单抗/伊匹木单抗）与单独使用伊匹木单抗进行成本效益分析（CEA），以探索I-O治疗反应和临床结果。将研究结果与传统的3状态PSM的结果进行比较。
该案例研究扩展了传统的3状态PSM，将进展前状态分为无反应者和反应者，将进展后状态分为正常和I-O进展，以考虑临床反应前的延迟治疗效果。使用患者层面的数据（尽可能）填充模型状态，从最佳总体反应（BOR）进行映射，并采用灵活和传统的参数方法进行生存分析。使用CheckMate 067试验的4年随访数据（研究时可获得的数据；此后已发表5年随访数据），将生存函数应用于各治疗组的无进展生存期（PFS）和总生存期（OS）终点。除了无进展和疾病进展的结果外，BOR信息还被用作区分I-O治疗反应的一种手段。采用英国国家医疗服务体系和个人社会服务（NHS/PSS）的终身视角，结果按每年3.5%进行贴现。
与传统的3状态PSM相比，5状态PSM在两个治疗组中均使质量调整生命年（QALY）增加，并对患者的健康状况进行了更细致的描述。使用5状态模型时，增量QALY增加了13%（从2.62个QALY增至2.95个QALY），增量成本降低了12%（从29,125英镑降至25,678英镑）。在两个模型中，该方案相对于单独使用伊匹木单抗的增量成本效益比（ICER）均在英国国家卫生与临床优化研究所（NICE）参考范围的下限内（每获得1个QALY为20,000英镑）。
纳入相关I-O健康状态的5状态经济模型，对于深入了解现有经济模型中通常未捕捉到的治疗反应和进展差异可能更具信息价值。包括与治疗反应相关的临床试验终点（这些终点在正在进行的I-O试验中未直接报告）可以映射到所提出的模型健康状态（尽管这样做需要一些假设）。未来I-O临床试验中治疗反应报告的改进有助于进一步验证和完善所提出的模型框架。
