Upregulation of thymidine phosphorylase (TP) by radiation (XRT): Phase II study of capecitabine (CAP) with XRT in pts with locally advanced (LA) pancreatic cancer

Abstract

14001 Background: Preliminary studies have shown that XRT unregulates TP, the main enzyme responsible for activity of CAP. The objectives of this phase II study were to further evaluate effect of XRT on TP, DPD, and TNF-alpha as well as response and efficacy of CAP with concurrent XRT in pts with LA pancreatic cancer. Methods: Pts received 50.4 Gy XRT with CAP at 1600 mg/m2 M-F × 6 wks determined from our phase I study (JCO, Dec 2005). Following CAP-XRT, stable and responding pts on CT scan were treated with CAP 2000mg/m2 × 14 days q 3 wks till progression. Restaging was performed every 9 wks. Tumor specimens were procured with EUS-FNA 1 wk prior and 2 wks after CAP-XRT to evaluate TP, DPD, and TNF-alpha mRNA levels by RT-PCR. A sample size of 20 was selected (mean 1 vs. S with mean 0=13.5; alpha = 0.05; power = .99; t-test). Results: 19 pts (median age: 67; M/F: 7/12) were enrolled at UAB between March 2004 and June 2005. 4 pts (21%) had confirmed partial responses and 13 (68%) had stable disease. 2 pts underwent surgery (Ro in 1; extensive fibrosis in 1). Six-month survival rate was 89%. We have not met enough deaths to estimate median survival. Grade 3 and 4 toxicities included: nausea/vomiting (5%), thrombosis (5%), hyperbilirubinemia (5%), and grade 3 GI bleeding (5%). No hematological toxicities except grade 1 thrombocytopenia (5%) and grade 1 anemia (10%). All pts completed full CAP-XRT with 3 dose reductions (by 20%). Pts received mean of 5.4 cycles (range: 3–15) of CAP alone with a total of 104 cycles with dose reductions in 7 cycles (by 25%). TP was elevated during wk 2 when compared to pre-XRT TP (P = .005) but DPD was not (P = .13) nor was TNF-alpha (P = .37). No correlation between TP and TNF-alpha was noticed. No association between TP/DPD ratio and efficacy of CAP was identified. Tumor TP expression was higher (183.16) in pt with GI bleeding. Conclusions: This Phase II study further confirms our Phase I results that CAP-XRT is an effective, tolerable, and an easy alternative to infusional 5-FU regimen for pts with LA pancreatic cancer. Also TP upregulation by XRT was statistically significant. While these results support use of CAP-XRT in pancreatic cancer, there appears to be additional genes (other than TP, DPD) associated with response to CAP and CAP-XRT. [Table: see text]