Others and we have demonstrated that hypoxia-inducible factor 1α (HIF-1α) and transcriptionally upregulated Aurora-A are required for disease progression in several tumors. We investigated the clinicopathological value of HIF-1α and Aurora-A in primary duodenal adenocarcinoma (PDA). Using immunohistochemistry, we evaluated Aurora-A and HIF-1α expression semiquantitatively in 140 PDA cases. There were 76 cases from one institute that formed the training set; 64 cases from another two institutes were used as the testing set to validate the prognostic value of Aurora-A and HIF-1α expression. Aurora-A expression was high or sufficient in the tumor zone, whereas expression was low in the adjacent normal epithelia. High Aurora-A expression, identified using the training set receiver operator characteristic (ROC) analysis-generated cutoff score, predicted poorer overall survival both in the testing set (18.0 vs. 45.1 %, P = 0.001) and training set (23.1 vs. 53.9 %, P = 0.011). Multivariate Cox regression confirmed that Aurora-A was an independent prognostic factor. Contrary to previous studies, we did not detect any correlation between Aurora-A and HIF-1α. Survival analysis showed that HIF-1α level was not correlated with patient outcome (P = 0.466). Activation of Aurora-A, an independent negative prognostic biomarker, might be used to identify particular PDA patients for more selective therapy.