COMPARISON OF TIMED AUTOMATA WITH DISCRETE EVENT SIMULATION FOR MODELING PERSONALIZED TREATMENT DECISIONS: THE CASE OF METASTATIC CASTRATION RESISTANT PROSTATE CANCER

Purpose: Traditional Markov Models are often inadequate to accurately represent complex and personalized treatment decisions over time. The aim of this study is to compare the usefulness of two promising alternative modeling techniques, Timed Automata (TA) originating from informatics, and Discrete Event Simulation (DES) known in operations research, for modeling these healthcare delivery processes involving multiple interactions and decision gates.

Method: The usefulness of both modeling techniques was assessed in a case study on the treatment of metastatic Castration Resistant Prostate Cancer (mCRPC). Circulating Tumor Cells (CTC) or cell-free tumor DNA may be used as a response marker for switching first (Docetaxel) to second line (Cabazitaxel) drug treatment. The use of these markers for early therapy switching was modeled using TA in UPPAAL and DES in Tecnomatix Plant Simulation. Techniques were compared on user-friendliness, input requirements and input possibilities, model checking facilities, and on actual outcomes for the case study. Input parameters were the same in both models and consisted of costs, QoL, treatment effectiveness, diagnostic performance, physicians’ behavior and survival. Primary outcome measures were costs, effectiveness expressed in QALYs and survival.

Result: Model building took several days for both techniques. Both modeling approaches yield comparable results. For TA, CTC reduced treatment with Docetaxel and Cabazitaxel by, on average,  108.9 and 107.6 days, respectively. For DES, treatment was reduced by 83.6 and 85.0 days. CTC therefore reduced healthcare costs by €28,998 and €21,992 according to TA and DES respectively.

While comparing the methods, it appeared that translating the process into a model was easier using TA, as this method allows independent modeling of the components comprising the treatment process such as patients, physicians, tests and treatments, whose mutual interaction and communication could be modeled easier and more extensive. Furthermore, the model checking feature of UPPAAL was found to be a powerful tool for validation of the model.

Conclusion: Timed Automata is a new and interesting alternative modeling technique, as it allows explicit separation of model components  and supports statistical model checking to validate models. Both Timed Automata and Discrete Event Simulation seem to be suitable for modeling complex and personalized treatment processes like that of metastatic Castration Resistant Prostate Cancer.