Introduction: Information on the relative cost-effectiveness of treatments for cancer is being increasingly sought as pressure on health care resources increases. The objective of this study was to assess the cost-effectiveness of gemcitabine/cisplatin (GC) versus cisplatin/etoposide (CE) in patients with advanced non-small cell lung cancer (NSCLC), using resource utilization data collected in conjunction with the first randomized clinical trial comparing both combinations.
Methods: Efficacy and medical care resource utilization data were collected prospectively in an open-label, multicenter, randomized, comparative, phase III trial conducted in Spain which compared gemcitabine/cisplatin and cisplatin/etoposide in 135 chemonaive patients with Stage IIIB or IV NSCLC. There were no differences between both regimens when survival was used as primary end-point, so a cost-minimization analysis was used to compare them. In addition, cost-effectiveness analyses were conducted when percentage of responses and time to progression were used as secondary end-points.
Results: There were no differences between both regimens when survival was selected as the efficacy end-point. Despite the higher chemotherapy cost of GC when compared to CE, there were no differences in total direct costs (584523 pts for GC and 589630 pts for CE; P=NS) between both regimens. Potential savings with GC were mainly associated with a decrease in hospitalization rate. There were differences in favor of GC when response rate (40.6% for GC and 21.9% for CE; P<0.05) and time to disease progression (8.7 months for GC and 7.2 months for CE; P<0. 05) were used as clinical end-points. GC presented a favorable cost-effectiveness profile when compared to CE.
Conclusions: This prospective economic evaluation conducted alongside a clinical trial offers valuable preliminary information on the potential efficiency of the combination gemcitabine-cisplatin in NSCLC. Future assessments based on larger clinical trials focused on survival and naturalistic economic studies conducted in real clinical practice settings are necessary to confirm these findings.