A pilot study to establish a clinical model to perform phase II studies of breast cancer chemopreventive agents in women at high risk with biomarkers as surrogate endpoints for activity

Abstract

Purpose: Use of surrogate end point biomarkers in phase II trials may help select agents that appear to have activity and might be evaluated in future phase III definitive trials of breast cancer prevention. We performed a pilot clinical trial to establish the logistics for a clinical model to perform phase II studies of breast cancer chemopreventive agents in women at high risk with novel imaging techniques and candidate surrogate end point biomarkers for activity. We chose tamoxifen to establish proof of principal with a known effective agent.

Experimental design: Women at a high risk of developing a new breast cancer and for whom tamoxifen was recommended were eligible. The women underwent baseline and 3 and 6 months mammogram and magnetic resonance imaging (MRI) of one breast to identify areas of water-like intensity (epithelial) and to determine the changes over time and MRI-directed core breast biopsies of these areas for surrogate end point biomarkers analysis.

Results: From August 1999 to March 2001, 26 women underwent baseline imaging and core biopsies. Sixteen women took tamoxifen and 10 chose not to. Overall, 79% of the samples contained glandular tissue evaluable for histology, but only 66% of the samples were evaluable for marker analysis. Only 12 patients had specimens with glandular tissue sufficient for marker analysis both at baseline and in at least one follow-up. Because of the small number of women with matched samples, marker analysis was not informative.

Conclusions: This study shows the feasibility of obtaining serial core breast biopsies from women at a high risk of developing a new breast cancer. Patient participation in this model is satisfactory, and such a model may provide indication of drug activity. MRI-directed biopsy did not provide a high yield of evaluable samples, and additional work on adequate collection of epithelial tissue for surrogate end point biomarker analysis is thus necessary.