Objectives: To evaluate the clinical and cost-effectiveness of capecitabine and tegafur with uracil (UFT/LV) as first-line treatments for patients with metastatic colorectal cancer, as compared with 5-fluorouracil/folinic acid (5-FU/FA) regimens.
Data sources: Electronic databases, reference lists of relevant articles and sponsor submissions were also consulted.
Review methods: Systematic searches, selection against criteria and quality assessment were performed to obtain data from relevant studies. Costs were estimated through resource-use data taken from the published trials and the unpublished sponsor submissions. Unit costs were taken from published sources, where available. An economic evaluation was undertaken to compare the cost-effectiveness of capecitabine and UFT/LV with three intravenous 5-FU/LV regimens widely used in the UK: the Mayo, the modified de Gramont regimen and the inpatient de Gramont regimens.
Results: The evidence suggests that treatment with capecitabine improves overall response rates and has an improved adverse effect profile in comparison with 5-FU/LV treatment with the Mayo regimen, with the exception of hand-foot syndrome. Time to disease progression or death after treatment with UFT/LV in one study appears to be shorter than after treatment with 5-FU/LV with the Mayo regimen, although it also had an improved adverse effect profile. Neither capecitabine nor UFT/LV appeared to improve health-related quality of life. Little information on patient preference was available for UFT/LV, but there was indicated a strong preference for this over 5-FU/LV. The total cost of capecitabine and UFT/LV treatments were estimated at 2111 pounds and 3375 pounds, respectively, compared with the total treatment cost for the Mayo regimen of 3579 pounds. Cost estimates were also presented for the modified de Gramont and inpatient de Gramont regimens. These were 3684 pounds and 6155 pounds, respectively. No survival advantage was shown in the RCTs of the oral drugs against the Mayo regimen. Cost savings of capecitabine and UFT/LV over the Mayo regimen were estimated to be 1461 pounds and 209 pounds, respectively. Drug acquisition costs were higher for the oral therapies than for the Mayo regimen, but were offset by lower administration costs. Adverse event treatment costs were similar across the three regimens. It was inferred that there was no survival difference between the oral drugs and the de Gramont regimens. Cost savings of capecitabine and UFT/LV over the modified de Gramont regimen were estimated to be 1353 pounds and 101 pounds, respectively, and over the inpatient de Gramont regimen were estimated to be 4123 pounds and 2870 pounds, respectively.
Conclusions: The results show that there are cost savings associated with the use of oral therapies. No survival difference has been proven between the oral drugs and the Mayo regimen. In addition, no evidence of a survival difference between the Mayo regimen and the de Gramont regimens has been identified. However, improved progression-free survival and an improved adverse event profile have been shown for the de Gramont regimen over the Mayo regimen. Further research is recommended into the following areas: quality of life data should be included in trials of colorectal cancer treatments; the place of effective oral treatments in the treatment of colorectal cancer, the safety mechanisms needed to ensure compliance and the monitoring of adverse effects; the optimum duration of treatment; the measurement of patient preference; and a phase III comparative trial of capecitabine and UFT/LV versus modified de Gramont treatment to determine whether there was any survival advantage and to collate the necessary economic data.