Issues in the Design of Ovarian Cancer Trials Evaluating Antineoplastic Agents Not Predicted to Cause Regression of Existing Tumor Masses

Despite the conduct of numerous randomized clinical trials in advanced ovarian cancer over the past decade, no new primary chemotherapeutic regimen has demonstrated a major improvement in overall survival (OS) compared with a program of a platinum agent plus a taxane. This suggests that a plateau in the effectiveness of cytotoxic chemotherapy regimens has been reached and that novel treatment paradigms are necessary. One approach is the use of maintenance therapy after an initial response to standard regimens has been achieved. This possibility is being explored with a number of targeted therapeutic strategies that focus on theorized unique molecular differences between tumors and normal tissue. By inhibiting or delaying tumor growth or spread, such agents delivered as maintenance therapy might extend the time to disease progression but might not necessarily be associated with improvements in objective response based on tumor shrinkage. Furthermore, these highly clinically relevant outcomes might not always be detectable in short-term trials oriented toward measurable tumor response, and longer treatment periods might be necessary to allow time for the identification of responder subsets or development of an immune response. Even defining the optimal regimen to undergo definitive testing could be problematic. As a result, innovative trial designs might be required to both detect potentially meaningful biologic activity and to identify patient subsets most likely to respond to treatment before largescale phase III trials are performed.