A new decision model for cost-utility comparisons of chemotherapy in recurrent metastatic breast cancer

Abstract

In the absence of comparative clinical and pharmacoeconomic trial data for docetaxel versus paclitaxel as second-line therapy for patients with anthracycline-resistant metastatic breast cancer, a computer-based decision-analysis model was designed to evaluate the comparative utility to patients of these two taxoids. The model used the Markov process to analyse disease states (response, stable disease, progressive disease) and toxicities (acute, cumulative) for each treatment during the period from commencement of up to six 3-weekly cycles of chemotherapy, to death. A cost-utility analysis was carried out using the model, with a probability, a cost and a utility determined for each health state. Response rates were obtained from clinical trial data supplemented by expert clinical opinion. Costs were taken from UK national databases and published sources and the published UK prices of docetaxel and paclitaxel. Utilities for the various health states were established by use of standard gamble and visual analogue methods assessed by 30 oncology nurses in the UK who were acting as proxy patients. The results of the model showed that response rate is the key parameter determining the utility and cost utility of treatments for metastatic breast cancer. Although the total per-patient cost associated with docetaxel was marginally higher than that for paclitaxel (8233 pounds vs 8013 pounds), the higher response rate associated with docetaxel produced an improvement in utility to the patient at an incremental healthcare cost that is acceptable according to available defined limits. Sensitivity analyses revealed that, although the model was sensitive to changes in response rate and drug costs, the cost-utility ratio for docetaxel versus paclitaxel varied within acceptable limits in response to all likely changes in key parameters. In summary, in the base case used in this model, docetaxel produces a substantially larger utility benefit than paclitaxel, at a small additional cost per QALY gained (equivalent to 7 pounds per additional day of perfect health).