Dynamic contrast-enhanced MRI perfusion for differentiating between melanoma and lung cancer brain metastases

Brain metastases originating from different primary sites overlap in appearance and are difficult to differentiate with conventional MRI. Dynamic contrast-enhanced (DCE)-MRI can assess tumor microvasculature and has demonstrated utility in characterizing primary brain tumors. Our aim was to evaluate the performance of plasma volume (Vp) and volume transfer coefficient (K^trans ) derived from DCE-MRI in distinguishing between melanoma and nonsmall cell lung cancer (NSCLC) brain metastases. Forty-seven NSCLC and 23 melanoma brain metastases were retrospectively assessed with DCE-MRI. Regions of interest were manually drawn around the metastases to calculate Vp_mean and Kmeantrans. The Mann-Whitney U test and receiver operating characteristic analysis (ROC) were performed to compare perfusion parameters between the two groups. The Vp_mean of melanoma brain metastases (4.35, standard deviation [SD] = 1.31) was significantly higher (P = 0.03) than Vp_mean of NSCLC brain metastases (2.27, SD = 0.96). The Kmeantrans values were higher in melanoma brain metastases, but the difference between the two groups was not significant (P = 0.12). Based on ROC analysis, a cut-off value of 3.02 for Vp_mean (area under curve = 0.659 with SD = 0.074) distinguished between melanoma brain metastases and NSCLC brain metastases (P < 0.01) with 72% specificity. Our data show the DCE-MRI parameter Vp_mean can differentiate between melanoma and NSCLC brain metastases. The ability to noninvasively predict tumor histology of brain metastases in patients with multiple malignancies can have important clinical implications.