Randomized evidence on chemotherapy and hormonal therapy regimens for advanced endometrial cancer: an overview of survival data.

Several chemotherapy and hormonal therapy regimens have been used in advanced endometrial cancer. In this review we have systematically evaluated the available data from randomized trials on survival. We searched MEDLINE, EMBASE and the Cochrane Library (last search April 2005) for randomized controlled trials evaluating various chemotherapy or hormonal therapy regimens in locally advanced or metastatic endometrial cancer. We focused on survival outcomes and examined trial characteristics pertaining to quality and potential biases. Across 17 eligible trials (2964 patients randomized), only 4 regimens were involved in more than one trial, and only two trials had used the same comparison of regimens. A statistically significant effect in survival was seen only in one recent trial, but it was borderline (P = 0.032) and amounted to only 3 months difference in median survival. Three more trials reported some survival benefits, but these were seen only after specific adjustments, and the difference was against the experimental arm in one of these three trials. Only four trials (24%) apparently analyzed all randomized patients and none of the trials were blinded. Median survival was seemingly longer in more recent compared with older trials, but this effect was driven by the inclusion of significantly fewer patients with poor performance status in more recent trials (P < 0.001). Overall, randomized evidence on systemic treatment in advanced endometrial cancer is fragmented and survival benefits for specific regimens are questionable.