OBJECTIVE: The purpose of this study was to determine whether combined diffusion-weighted imaging and dynamic susceptibility contrast-enhanced perfusion-weighted imaging magnetic resonance imaging can be used to differentiate between common malignant brain tumors, including lymphomas and high-grade gliomas. METHODS: We evaluated 87 patients with histologically confirmed brain tumors, including 33 primary central nervous system lymphomas (PCNSLs) and 54 glioblastomas (GBMs). All patients underwent conventional magnetic resonance imaging, diffusion-weighted imaging, and perfusion-weighted imaging before surgical removal of the lesion or stereotactic biopsy. RESULTS: The maximum relative cerebral blood volume (rCBV) ratios of GBMs were significantly higher than those of PCNSLs (P < 0.0001). The maximum rCBVs helped to distinguish PCNSLs from GBMs with 97.0% sensitivity, 90.7% specificity, and 0.98 area under the curve. The minimum apparent diffusion coefficients (ADCs) of PCNSLs were significantly lower than those of GBMs (P < 0.0001). At an rCBV cutoff value of 4.0 and a minimum ADC of 1.0 × 10-3 mm2/second, it was possible to differentiate between PCNSLs and GBMs. CONCLUSIONS: The combination of rCBV and ADC can facilitate the differentiation between PCNSLs and GBMs.