OBJECTIVES: To develop a decision-analytic model describing the current course of disease, including the available treatments, in metastatic colorectal cancer to serve as a baseline for the analyses of new developments. METHODS: A disease model was constructed containing six disease states, namely  ‘first-line treatment’, ‘progression of disease after first-line’,’ second-line treatment’, ‘progression of disease after second-line’, ‘third-line treatment’ and ‘death’. Time spent in each state was predicted using log-logistic, log-normal or Weibull survival models, utilizing simulated characteristics. All survival models and simulated characteristics were based on patient-level data, provided by the CAIRO trial (n=820). The model was validated by comparing various outcome parameters with the original data.  RESULTS: The overall survival of the simulated cohort resulted in a median survival of 498 days. The original patient level data had a median OS of 508 days (95% CI: 463- 545).    For the subgroups age and treatment line data continued to match consistently with patient-level data. Patients treated with sequential therapy in the RCT had an OS of 492 days (95%CI 416-557), the model predicted an OS of 489 days. Patients in the RCT with combination therapy had an OS of 525 days (95%CI 442-591days), the model predicted 509 days OS.  Similar results were achieved in other subgroups and for various other outcome measures. CONCLUSIONS: The model constructed during this study is a good basis on which to model the effectiveness and cost-effectiveness of numerous new treatment options, including expensive drugs. It will also allow - in future - for the evaluation of these treatment options while accounting for the heterogeneous nature of colorectal cancer As such  this model  offers a method to aid in determining the optimal sequence of  treatments for patients with metastatic colorectal cancer, both in terms of costs and effects.