SECOND-LINE TREATMENT, PACLITAXEL (MEDIAN PROGRESSION-FREE SURVIVAL): The median progression-free survival in the paclitaxel arm was 3.5 months. This was significantly longer than the mitomycin control arm (1.6 months, p = 0.026). BREAST CANCER - SECOND-LINE TREATMENT, PACLITAXEL (MEDIAN OVERALL SURVIVAL): The median length of overall survival in the paclitaxel arm was 12.7 months, compared with 8.4 months in the mitomycin arm. BREAST CANCER - SECOND-LINE TREATMENT, PACLITAXEL (QUALITY OF LIFE): Quality of life was not reported. BREAST CANCER - SECOND-LINE TREATMENT, PACLITAXEL (ECONOMIC EVALUATION): The only economic evaluation that compared paclitaxel with control (mitomycin) was submitted in confidence and has been removed from this report. Six economic evaluations involved comparisons of paclitaxel and docetaxel, which are given below. BREAST CANCER - SECOND-LINE TREATMENT, DOCETAXEL: Four randomised controlled Phase III trials were identified: 303 Study, 304 Study, Scand and Bonneterre. A total of 1092 patients were included. One of these was a preliminary report of a study before completion of accrual (Bonneterre). Patients in the 303 Study had previously received chemotherapy involving alkylating agents; those in the other three had received anthracyclines. There were six economic evaluations on docetaxel. BREAST CANCER - SECOND-LINE TREATMENT, DOCETAXEL (QUALITY OF TRIALS): The 303 and 304 Studies were analysed on an intention to treat basis; the Scand trial excluded a single patient. The length of follow-up ranged from 11 months (Scand) to 23 months (303 Study). At least two-thirds of the participants in these trials had died. The Scand study recommended cross-over to alternate treatment on objective signs of disease progression. Patients crossing over in this way were violating the randomisation; however, no details were given concerning whether or not such patients were censored. In the economic analyses, there were no direct comparisons for the estimation of benefits. BREAST CANCER - SECOND-LINE TREATMENT, DOCETAXEL (MEDIAN PROGRESSION-FREE SURVIVAL): The median progression-free survival in the docetaxel arm ranged from 4.75 months (304 Study) to 7 months (Bonneterre). Patients in the docetaxel arms of the 304 and Scand studies had significantly longer progression-free survivals than controls (4.75 months versus 2.75 months, p = 0.001; 6.3 months versus 3 months, p = 0.001). BREAST CANCER - SECOND-LINE TREATMENT, DOCETAXEL (MEDIAN OVERALL SURVIVAL): The median overall survival in the docetaxel arm ranged from 10.4 months (Scand) to 15 months (303 Study). Patients in the docetaxel arms of the 304 Study survived for significantly longer than the mitomycin plus vinblastine arm (11.4 months versus 8.7 months, p = 0.03). BREAST CANCER - SECOND-LINE TREATMENT, DOCETAXEL (QUALITY OF LIFE): Quality of life was evaluated in two of the trials: the 303 and 304 Studies. There were no significant differences between docetaxel and control in either of these trials in terms of global health status, although differences were apparent on some subscales. These did not appear to follow a consistent pattern across the trials. BREAST CANCER - SECOND-LINE TREATMENT, DOCETAXEL (ECONOMIC EVALUATIONS): All six of these involved comparisons of paclitaxel and docetaxel, where the range of cost-utility ratios for incremental quality-adjusted life-years (QALYs) gained was pound 1990-pound 2431. In addition, three analyses compared docetaxel and vinorelbine. The cost-utility ratio for incremental QALYs gained was pound 14,050 in the only one of these carried out in the UK. OVARIAN CANCER - FIRST-LINE TREATMENT, PACLITAXEL: Four randomised controlled Phase III trials were identified: EORTC, TITGANZ, E1193 and CA139-278. (ABSTRACT TRUNCATED)