Use of changes in CA125 doubling time to detect activity of cytostatic agents in women relapsing with ovarian carcinoma. Study 1 - tamoxifen

Abstract

5119 Background: Many new targeted-therapy agents with potential anti-cancer activity could be cytostatic rather than cytotoxic. Therefore standardised radiological response criteria such as RECIST may not detect activity of these agents in phase II trials. There is an urgent need for a method to screen for activity of these agents before embarking on large randomised trials. The main purpose of this study was to determine whether there is a predictable log-linear rise of CA125 levels in asymptomatic relapsing ovarian cancer patients and if so, whether a change in rate of rise of CA125 can be used to demonstrate activity of a cytostatic agent. For proof of principle, we used tamoxifen in this trial which has modest clinical activity, and so one would expect some changes in CA125 doubling time. Methods: Ovarian cancer patients with previously elevated CA125 levels that had successfully completed first relapse chemotherapy were eligible. Patients consented to monthly CA125 measurements, which became fortnightly once there was any rise in levels. Tamoxifen 20mg once daily was commenced once CA125 was either four times its upper limit of normal, or four times the nadir level if this was elevated at baseline. Log CA125 levels were plotted to look for linear trends. Results: So far out of the first 15 patients that have entered the study, 12 have had rising CA125 levels, out of whom 6 patients demonstrated a log-linear rise. The median number of samples from the start of the linear rise to start of tamoxifen was 5. A change in the slope of log linear CA125 levels was observed in 2 out of 6 patients started on Tamoxifen. Conclusions: Early results suggest that this approach to assessing new cytostatic agents in phase II trials does look feasible. A log linear rise was seen in 50% of patients. We aim to recruit 200 patients to reliably assess whether Tamoxifen has activity based on a change in slope of CA125. No significant financial relationships to disclose.