Thymidilate synthase (TS), thymidine phosphorylase (TP), dihydropyrimidine dehydrogenase (DPD), and dihydrofolate reductase (DHFR) as predictive markers of capecitabine efficacy in breast cancer patients

11033 Background: The efficacy of 5-FU and the pro-drug capecitabine (C) is mediated through the thymidylate synthase pathway and may be affected by expression of key enzymes in this pathway. The expression and/or gene copy number of TS, TP, DPD and DHFR was assessed and correlated with time-to progression (TTP) and progression-free survival (PFS). Methods: Adult female patients with pathologically confirmed breast cancer and locally advanced or metastatic disease were treated with C 1000 mg/m2 BID days 1–14 of a 21-day cycle. Formalin-fixed, paraffin-embedded (FFPE) specimens from C-treated patients were processed for RNA extraction. Total RNA was isolated from 32 FFPE tissue samples containing at least 70% tumor cells and RNA levels for TS, TP, DPD and DHFR were quantified using real time RT-PCR and Affymetrix GeneChip microarrays. Custom made TS and TP FISH probes (Dako, Glostrup, Denmark) were used to evaluate gene copy number and gene to reference ratios in at least 60 morphologically intact non-overlapping nuclei. Markers were correlated with TTP and PFS using Cox proportional hazard models in 24 patients. Results: Higher TS gene copy number was significantly associated with a decrease in PFS (HR 1.46, 95% CI 1.08 to 1.96, p=0.014) and TTP (HR 1.49, 95% CI 1.05 to 2.13, p=0.028). However, the association between TS RNA levels (expressed as Ct values) and PFS (HR 0.82, 95% CI 0.61 to 1.11, p=0.198) or TTP (HR 0.74, 95% CI 0.49 to 1.14, p=0.172) failed to reach statistical significance. RNA levels, determined by Affymetrix, were significantly correlated with RT-PCR for TS (r= -0.5073, p=0.0004) and DHFR (r=-0.50994, p=0.003). Neither TP gene copy number nor RNA levels were significantly associated with PFS (HR 099, and HR 0.93) or TTP (HR 1.07 and HR 1.029), respectively. The association between RNA levels and PFS or TTP did not reach statistical significance for both DPD (HR 1.17 and HR 1.13) and DHFR (HR 0.68 and HR 0.63), respectively. Conclusions: These data indicate that TS gene copy number, assessed by FISH with proper standardization, might be a useful and easily accessible marker for C sensitivity in human breast cancer and warrants further investigation. [Table: see text]