The feasibility of using ultrasound for transcranial imaging and targeted drug delivery in the brain

Kullervo Hynynen1,2
1Department of Medical Biophysics, University of Toronto, Toronto, Ontario, Canada
2Physical Sciences Platform, Sunnybrook Research Institute, Toronto, Ontario, Canada

ABSTRACT
Recent development in ultrasound methods have made noninvasive focusing of ultrasound into deep targets in the brain possible.  This method is now used under Magnetic Resonance Imaging for the treatment of  essential tremor, chronic pain and Parkinson's disease. However, the method when combined with standard diagnostic ultrasound imaging  contrast agents containing micron sized gas bubbles, has shown great promise for focal opening of the Blood-Brain barrier (BBB) for targeted drug and cell delivery. The bubbles that are injected in the blood stream act as energy concentrators and insert significant physical stretching on the capillary walls when the bubbles expand and contract with the pressure wave. Therefore very low time average powers (in the order of a few mWs) are deposited in the brain thus having minimal impact on brain tissue.  The BBB heals itself in approximately in 6 h but is dependent on the amplitude of the acoustic pressure.  A wide variety of therapeutic agents have been delivered in the brain in animals ranging from chemotherapy agents to antibodies to stem cells. The treatments have been tested in disease models and found effective against for example brain tumours and Alzheimer's Disease.  It is expected that the clinical testing of this method may start in the near future.

The skull aberration correction methods and the multi-element ultrasound phased arrays that have made through skull sonications possible can be used also for brain imaging with high resolution. The ultrasound excited microbubbles also emit ultrasound that can be detected outside of the skull with sensitive receivers. By using a large number of receivers distributed all around the skull, the bubbles can be localized.  The localization of sound sources has been used in acoustics and is called Passive Acoustic Mapping (PAM). By incorporating skull-specific aberration corrections into a conventional PAM algorithm allows microbubble echoes to be imaged through the skull. By utilizing super-localization methods originally developed for optical microscopy, high resolution images of the bubbles can be created though the skull.  The phantom images obtained through a human skull show promise and indicate that the method may eventually be used for the mapping of the complete brain vasculature.