BACKGROUND: This study utilized the imaging data of primary liver cancer (PLC) treated with floxuridine (FUDR) and bevacizumab to test the hypothesis that dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) parameters correlate with tissue hypoxia markers and treatment outcome. METHODS: Seventeen patients with PLC were treated with hepatic artery infusional (HAI) FUDR for 14 days followed by systemic bevacizumab therapy. DCE-MRI images were obtained at baseline and after HAI FUDR and bevacizumab therapy. The parameters (K(trans), AUC) pertaining to perfusion and vascular permeability of the tumor and adjacent liver parenchyma were measured with DCE-MRI. Tissue obtained at baseline was stained for hypoxia markers (anti-hypoxia inducible factor-1α, anti-carbonic anhydrase IX, and vascular endothelial growth factor). Changes in DCE-MRI parameters were correlated with tissue hypoxia and time to progression (TTP). RESULTS: The median TTP was 8.8 months. Significant decreases in AUC90 (P = 0.004), AUC180 (P = 0.004), and K(trans) (P = 0.05) were noted in tumors after bevacizumab but not in nontumor areas. TTP correlated inversely with changes in AUC90 and AUC180 after bevacizumab (P = 0.002 and P = 0.0001). Reductions in tumor perfusion (AUC90 and AUC180) were greater in tumors expressing anti-hypoxia inducible factor-1α (P = 0.02 and 0.03), vascular endothelial growth factor (P = 0.01 and P = 0.01), and anti-carbonic anhydrase IX (P = 0.009 and P = 0.009). CONCLUSIONS: In patients with PLC, bevacizumab induces a reduction in tumor perfusion measured by DCE-MRI. These changes correlate with TTP and tissue markers of tumor hypoxia.