Abstract P2-09-24: Relationship of Capecitabine Response to Thymidine Phosphorylase Expression and Overall Survival in Advanced Breast Cancer

Abstract

Abstract Background: Capecitabine is an oral fluoropyrimidine carbamate that is metabolised to FU through three enzymatic steps. Thymidine phosphorylase (TP) is involved in the final step of the conversion from 5'-deoxy-5-fluorouridine (5'-DFUR) to 5-FU. TP is highly expressed in cancer cells compared to normal tissue. By exploiting this difference in TP activity, capecitabine is activated and preferentially converted into 5-FU within tumours. Besides this, TP is identical to platelet-derived endothelial cell growth factor (PD-ECGF), a known angiogenic factor. While TP expression is predictive of benefit from capecitabine in colorectal cancer (CRC), initial studies in advanced breast cancer (ABC) examining the predictive value of TP and related enzymes have produced mixed results. Furthermore, whether response to capecitabine is associated with any overall survival (OS) benefit has not been studied. Materials and Methods: This retrospective study analysed 92 ABC patients who received at least one cycle of capecitabine for locally advanced or metastatic disease with intention to treat. Best responses to capecitabine (assessed clinically or by imaging) whilst patients were on capecitabine were included for analysis. TP expression was analysed by immunohistochemistry in primary excised samples. Composite TP scoring (including cytoplasm and stroma) was used to define TP status. Chi-squared tests were carried out to find associations between TP and clinical factors. Kaplan Meier survival curves were produced for relapse free survival and OS. The logrank statistic was calculated to test independence of the survival curves by category. Results: On multivariate analysis, when the TP composite factor was added to a Cox regression model for OS using age, ER, grade, nodal status and size, it appeared to act as an independent prognostic factor for OS after first primary (hazard ratio: 0.45, p=0.01). Overall, responses included partial response (PR) in 38% and stable disease (SD) for 3 months or longer in 25% patients. There was no statistically significant correlation for capecitabine response and TP status. Also there was no significant difference in time to first relapse or from first relapse to start of capecitabine stratified by response, showing that response was not intrinsically related to better prognosis or slower growing cancers. However, there was a significantly prolonged OS equivalent for both SD and PR to capecitabine (logrank, p=0.02). Discussion: Most studies in ABC have not found high TP expression alone to predict better response to capecitabine. The results of our study, the largest to date, confirm these findings. A combined approach studying plasma levels of 5'-DFUR along with intra-tumoral TP should be investigated further. Interestingly, the finding of high TP as an independent prognostic factor could be due to deprivation of thymidine needed for proliferation, but this also needs further investigation. By analysing the long term course of breast cancer based on ultimate pattern of response, this study suggests a substantial benefit of capecitabine even though second or third line. It also suggests that if patients achieve SD therapy should continue. Citation Information: Cancer Res 2010;70(24 Suppl):Abstract nr P2-09-24.