In-vivo longitudinal MRI study: an assessment of melanoma brain metastases in a clinically relevant mouse model.

Brain metastases are an important clinical problem. Few animal models exist for melanoma brain metastases; many of which are not clinically relevant. Longitudinal MRI was implemented to examine the development of tumors in a clinically relevant mouse model of melanoma brain metastases. Fifty thousand human metastatic melanoma (A2058) cells were injected intracardially into nude mice. Three Tesla MRI was performed using a custom-built gradient insert coil and a mouse solenoid head coil. Imaging was performed on consecutive days at four time points. Tumor burden and volumes of metastases were measured from balanced steady-state free precession image data. Metastases with a disrupted blood-tumor barrier were identified from T1-weighted spin echo images acquired after administration of gadopentetic acid (Gd-DTPA). Metastases permeable to Gd-DTPA showed signal enhancement. The number of enhancing metastases was determined by comparing balanced steady-state free precession images with T1-weighted spin echo images. After the final imaging session, ex-vivo permeability and histological analyses were carried out. Imaging showed that both enhancing and nonenhancing brain metastases coexist in the brain, and that most metastases switched from the nonenhancing to the enhancing phenotype. Small numbers of brain metastases were enhancing when first detected by MRI and remained enhancing, whereas other metastases remained nonenhancing to Gd-DTPA throughout the experiment. No clear relationship existed between the permeability of brain metastases and size, brain location and age. Longitudinal in-vivo MRI is key to studying the complex and dynamic processes of metastasis and changes in the blood-tumor barrier permeability, which may lead to a better understanding of the variable responses of brain metastases to treatments.