A DISCRETE EVENT SIMULATION MODEL FOR RENAL CELL CARCINOMA FOR USE IN A COST-EFFECTIVENESS ANALYSIS

Author(s)

Dansk V1, Treharne C2, Santi I2, Malcolm B3, Johal S2
1PAREXEL, Stockholm, Sweden, 2Parexel International, London, UK, 3Bristol-Myers Squibb, Uxbridge, UK

OBJECTIVES: To develop a flexible and comprehensive discrete event simulation model (DES) for nivolumab versus everolimus in the treatment of renal cell carcinoma (RCC) that captures differences in treatment outcomes and costs between patients with heterogeneous baseline characteristics.

METHODS: A DES was developed in Microsoft Excel based on individuals experiencing three key events: disease progression, treatment discontinuation, and death. Risk prediction equations were derived from 24-month patient-level data from the CheckMate 025 trial, using a step-wise backward elimination process to identify relevant predictors of the risk of progression or death. Hypothetical patients were generated from a multivariate normal distribution based on the characteristics of patients enrolled in CheckMate 025. Times to event for individuals were estimated from the predictive equations for overall survival, progression-free survival and post-progression survival. The survival curves chosen were based on comparison of the Akaike information criterion. The model accounted for differences in survival between patients who continued on their allocated treatment beyond progression and those who did not.

RESULTS: Four predictive equations were derived from the trial data. The following baseline characteristics were identified as significant predictors of survival; age, Karnofsky performance status, haemoglobin, time from diagnosis to randomisation and tumour size. Mean overall survival of 53.0 months and 34.8 months were estimated for nivolumab and everolimus, respectively, with mean post-progression survival of 43.0 and 26.3 months, respectively.

CONCLUSIONS: The DES provides a flexible approach to capture patient heterogeneity and history when predicting long-term disease progression and survival outcomes for use in a cost-effectiveness analysis, and may be applied to other oncology settings. This becomes important in the context of immunotherapies with a novel mechanism of action and where extrapolation beyond trial duration requires consideration of such issues. Standard modelling approaches, such as the partitioned survival model, may not completely address these challenges.

Conference/Value in Health Info

2017-11, ISPOR Europe 2017, Glasgow, Scotland

Value in Health, Vol. 20, No. 9 (October 2017)

Code

PRM100

Topic

Methodological & Statistical Research

Topic Subcategory

Modeling and simulation

Disease

Oncology

Explore Related HEOR by Topic


Your browser is out-of-date

ISPOR recommends that you update your browser for more security, speed and the best experience on ispor.org. Update my browser now

×