T2 Imaging in Monitoring of Intraparenchymal Real-Time Convection-Enhanced Delivery

BACKGROUND: Real-time convection-enhanced delivery (RCD) of adeno-associated viral vectors by co-infusion of gadoteridol allows T1 magnetic resonance imaging (T1 MRI) prediction of areas of subsequent gene expression. The use of T2 MRI in RCD is less developed. In addition, the effect of flushing a dead-space volume on subsequent distribution of a therapeutic agent is not known. OBJECTIVE: The value of T2 MRI in RCD was investigated by comparing distribution volumes of saline with immediately after T1 RCD of gadoteridol and by comparing T2, T1, and transgene distribution patterns after viral vector RCD. METHODS: Adult nonhuman primates underwent saline infusion/T2 acquisition, immediately followed by gadoteridol infusion/T1 acquisition in the putamen and brainstem. Distribution volumes and spatial patterns were analyzed. Gadoteridol and adeno-associated virus encoding human aromatic l-amino acid decarboxylase (AAV2-hAADC) were co-infused under alternating T2/T1 acquisition in the thalamus, and hyperintense areas were compared with areas of subsequent transgene expression. RESULTS: Ratios of distribution volume to infusion volume were similar between saline and gadoteridol RCD. Spatial overlap correlated well between T2 and T1 images. The second infusate followed a spatiotemporal pattern similar to that of the first, filling the target area before developing extra-target distribution. Areas of human L-amino acid decarboxylase expression correlated well with areas of both T1 and T2 hyperintensity observed during RCD. CONCLUSION: Accuracy of cannula placement and initial infusate distribution may be safely determined by saline infusion without significantly altering the subsequent distribution of the tracer agent. T2 RCD provides detection of intraparenchymal convection-enhanced delivery in the uninjured brain and may predict subsequent distribution of a transgene after viral vector infusion.