Introduction: Neoadjuvant chemoradiation (NCR) has become the preferred approach for locally advanced rectal cancer (RC). We have evaluated the use of oral topotecan as a radiosensitizing agent in this setting. In this study, by the use of gene expression profiling (GEP), we sought to assess molecular patterns associated with pathologic response to NCR. Methods: A phase I study design was used to determine the maximum tolerated dose of oral topotecan in combination with radiation in the neoadjuvant treatment of RC. Topotecan was administered orally for 25 doses with a starting dose of 0.25mg/m2 (max 1.15 mg/m2) in combination with standard radiation. All patients underwent tumor biopsy both before and during NCR treatment (between days 10-14). Pathologic response was determined after definitive radical surgery. All specimens underwent GEP using the Affymetrix U133 Plus 2.0 Gene chip and data were normalized using the Bioconductor software. Significant genes were identified by the t test. GeneGO MetaCore software was used to identify significantly altered biologic pathways. Results: From 2001 to 2007, 26 patients (Stage II n=18, III n=8) participated in the study. The pathologic complete response (pCR) and non-response rate (NR, stable or progressive disease) were 19% (n=5) and 39% (n=10), respectively. A partial pathologic response was observed in 42% of patients (n=11). Comparing pre-treatment GEP between pCR and NR patients revealed 30 upregulated and 37 downregulated genes (p<0.001). Gene ontology analysis revealed 7 significantly enriched pathways. Of these, 6 were involved in nucleotide or lipid metabolism (e.g. CTP/UTP metabolism, p<0.001; plasmalogen metabolism, p<0.001) and 1 in apoptosis and survival (antiapoptotic TNFs/NF-kB/IAP pathway, p=0.07). In an attempt to identify mid-treatment markers of subsequent response, we examined gene expression ratios from pre- and intra-treatment tissues. A total of 18 genes demonstrated differential upregulation in pCR cases and included genes associated with angiogenesis, apoptosis and transcriptional regulation (e.g. ANGPT1, NR1D1, LOC10012, EYA4 and PDCD2). Conclusions: Neoadjuvant topotecan and radiation yields pCR rates similar to those seen with standard 5-FU/radiation. Distinct molecular patterns were detectable in pretreatment biopsies that were associated with subsequent pathologic response. As a novel approach, we have also identified specific molecular changes that occur during treatment that may be predictive of a pCR. Our findings encourage further investigation of predictive genetic signatures as well as of biologic mechanisms associated with NCR treatment response.(Quickshot/Poster)