PURPOSE: Tumor-related cancer pain often comprises mixed pain with both nociceptive and neuropathic components. Whether tumor-related cancer pain includes a neuropathic component impacts the therapeutic strategy. The aim of this cross-sectional study was to investigate the usefulness of two screening tools for neuropathic pain, painDETECT and Self-Report Leeds Assessment of Neuropathic Symptoms and Signs (S-LANSS), in identifying the neuropathic component of mixed pain among patients with tumor-related cancer pain. METHOD: This cross-sectional study recruited consecutive inpatients and outpatients at a single site. The diagnostic accuracy of painDETECT and S-LANSS was evaluated using receiver operating characteristic curve analysis and classification probability. RESULTS: Of the study group, 106 patients had tumor-related cancer pain. Analyses of the nociceptive and mixed pain groups (n = 104) showed that neither painDETECT nor S-LANSS had satisfactory areas under the curve (AUCs) for identifying the neuropathic component of mixed pain (0.59 for painDETECT and 0.56 for S-LANSS). By pain intensity, the AUC for painDETECT was significantly higher in the mild pain group than in the moderate or severe pain group (0.77 vs. 0.43, P = 0.002). All parameters of classification probability for both tools were higher in the mild pain group than in the moderate or severe pain group. CONCLUSIONS: painDETECT and S-LANSS could not identify the neuropathic component of mixed pain among patients with tumor-related cancer pain, especially when pain was moderate or severe. Contrarily, these screening tools might be useful for identifying the neuropathic component of mixed pain for mild pain.