Prediction of radiosensitivity in primary central nervous system germ cell tumors using dynamic contrast-enhanced magnetic resonance imaging.

OBJECTIVE To evaluate the feasibility of dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) for predicting tumor response to radiotherapy in patients with suspected primary central nervous system (CNS) germ cell tumors (GCTs). METHODS DCE-MRI parameters of 35 patients with suspected primary CNS GCTs were obtained prior to diagnostic radiation, using the Tofts and Kermode model. Radiosensitivity was determined in tumors diagnosed 2 weeks after radiation by observing changes in tumor size and markers as a response to MRI. Taking radiosensitivity as the gold standard, the cut-off value of DCE-MRI parameters was measured by receiver operating characteristic (ROC) curve. Diagnostic accuracy of DCE-MRI parameters for predicting radiosensitivity was evaluated by ROC curve. RESULTS A significant elevation in transfer constant (K(trans)) and extravascular extracellular space (Ve) (P=0.000), as well as a significant reduction in rate constant (Kep) (P=0.000) was observed in tumors. K(trans), relative K(trans), and relative Kep of the responsive group were significantly higher than non-responsive groups. No significant difference was found in Kep, Ve, and relative Ve between the two groups. Relative K(trans) showed the best diagnostic value in predicting radiosensitivity with a sensitivity of 100%, specificity of 91.7%, positive predictive value (PPV) of 95.8%, and negative predictive value (NPV) of 100%. CONCLUSIONS Relative K(trans) appeared promising in predicting tumor response to radiation therapy (RT). It is implied that DCE-MRI pre-treatment is a requisite step in diagnostic procedures and a novel and reliable approach to guide clinical choice of RT.