IMPORTANCE OF MULTIAGENT CHEMOTHERAPY REGIMENS IN OVARIAN-CARCINOMA -DOSE INTENSITY ANALYSIS

Background: In the previous meta-analysis of dose intensity (dosage) o f chemotherapy in advanced ovarian cancer, we analyzed data on cycloph osphamide, altretamine (hexamethylmelamine), doxorubicin, and cisplati n. Only cisplatin showed statistically significant association of comp lete and partial clinical response with dose intensity. Purpose: This analysis updates the previous results and further characterizes respon se to cisplatin alone or in multiagent regimens. Methods: We analyzed data from 18 regimens containing platinum (cisplatin or carboplatin) t hat were used in nine new randomized trials, in addition to data from the 60 groups of patients in our previous study in which responses wer e reported. Relative dose intensity was calculated as a fraction of th e dosage of a drug in the standard regimen of cyclophosphamide, altret amine, doxorubicin, and platinum (CHAP). We performed single and multi ple regression analyses to determine the relationship between disease outcome and relative dose intensity for cyclophosphamide, platinum, an d doxorubicin alone or in combination. Results: The association betwee n outcome and dose intensity for platinum alone or in multiagent regim ens was statistically significant. This association was of borderline significance for cyclophosphamide alone but was not significant for th is drug in multiagent regimens. There were insufficient data to test t he relationship for doxorubicin as a single agent, but in multiagent r egimens, the relationship was borderline (P = .05). Multiagent regimen s containing platinum produced greater response rates than platinum al one for any fixed, planned relative dose intensity for platinum. Concl usions: Our results support other published findings that use of cyclo phosphamide and doxorubicin increases the efficacy of single-agent pla tinum. Relative dose intensity values for cyclophosphamide used alone were larger than those used in multiagent regimens, which might explai n why the relationship between relative dose intensity and outcome for cyclophosphamide was not significant for use in multiagent regimens. Similarly, none of the multiagent regimens incorporated doxorubicin at a relative dose intensity for which the drug is found to be effective as a single agent. Implications: Prospective clinical trials are requ ired to test the effect of higher relative dose intensity for doxorubi cin and cyclophosphamide added to platinum in advanced ovarian cancer. An important element in the design of prospective trials will be to t est for the relative importance of dose intensity versus total dose. T his testing is best achieved in a three-arm study design such as that reported in adjuvant treatment of stage II breast cancer conducted by the Cancer and Leukemia Group B.