MRI Guided Focused Ultrasound in Breast Cancer Treatment

Institution: Stanford University
Investigator(s): Rachel Bitton, Ph.D. -
Award Cycle: 2010 (Cycle 16) Grant #: 16FB-0090 Award: $88,257
Award Type: Postdoctoral Fellowship
Research Priorities
Detection, Prognosis and Treatment>Imaging, Biomarkers, and Molecular Pathology: improving detection and diagnosis



Initial Award Abstract (2010)

Improved technologies for breast cancer screening have led to an increased detection of tumors that cannot be felt through palpation. Because these tumors are non-palpable, the current method to locate them for breast conservation surgery is image guided, single wire placement. However, wire localization lacks precision. It can result in positive tumor margins, which is the largest cause of local recurrence after breast conservation surgery, in 14-57% of patients. A new approach for pre-operative localization is to use MR-Image guided Focused Ultrasound (MR-FUS). By using a highly focused ultrasound beam, a small (2mm x 8mm) thermal hot spot kills cells deep within the body, while completely sparing healthy adjacent tissues. The mechanical properties of the tissue change due to heating, and can become stiffer. We hypothesize that the thermal lesions created by MR-FUS can precisely mark the perimeter non-palpable tumors, even irregularly shaped tumors, so that surgeons can feel where the tumor is during surgery.

To further advance the MR-FUS approach, the present study will develop a MR Acoustic Radiation Force Imaging (MR-ARFI) sequence that uses non-thermal ultrasound pulses to displace tissue. The MR-ARFI sequence will be used to 1) calibrate the ultrasound focal spot location in the tissue and to 2) monitor the focal spot during the heating phase of ablation. It will also be used to measure changes in tissue displacement, which correlate to a change in tissue stiffness of the ablated lesions. Next, we will optimize the ultrasound parameters. Thermal lesions will be tested separately in various tissue types found in the breast: fatty, primarily fibroglandular, and heterogeneous spots with both fatty and fibroglandular components. Lastly, the MR-FUS method will be compared against the wire localization method. Non-palpable mock tumors with MR contrast agent will be injected into cadaver breasts. Half of the breast specimens will receive pre-operative image guided tumor localization using the wire localization technique, and the other half will be localized using the MR-FUS method. A trained breast surgeon will use the localization methods to excise the mock tumor from the breast. The results of each method will be compared by imaging the remaining breast tissue to evaluate tumor margins.

This research is aimed to better breast conservation surgery with an image guided intervention. By changing the approach for breast conservation surgery, this project has the potential improve surgical outcomes, and to reduce the heavy patient burden of repeat surgeries.




Final Report (2012)

The purpose of this project was to develop and validate a new method for pre-surgical breast tumor marking in human breast tissue using MRI guided focused ultrasound (MR-FUS). MR-FUS could be used to draw a palpable boundary of lesions around a non-palpable breast tumor, providing a surgical guide.

In the first phase of the project we assessed feasibility of the technique. We were able to successfully create MR-FUS lesions of an arbitrary shape in cadaveric breast tissue with no known pathology. A key issue for the technique is to validate the change of palpability in MR-FUS lesions. We aimed to develop a method that could test the efficacy of palpability change of a lesion during the treatment, and using no more equipment, or repositioning the patient. Lesion validation would allow for on-the-fly remarking of undertreated lesions. We created a relative stiffness technique using MR-ARFI (Acoustic Radiation Force Imaging). Displacement maps showed changes in lesion stiffness of both mixed fibrogladular and adipose breast tissue. We optimized a ultrasound protocol that used stacked sonications to provide greater palpability in larger breasts, where the tumor boundary is deeper within the breast.

The final phase of the project was to conduct a comparison study of excision margins between wire localization and MR-FUS localization. Without conducting a phase I clinical trial in humans, the comparison would take place in cadaveric breast tissue with no known pathology, where a breast surgeon would excise implanted non-palpable tumors. This aim was discontinued due to the lack of completely non-palpable, injectable mock tumor. Any palpability of mock tumors in the cadaveric breast biases surgical excision results. This finding steered the project towards preparing for the next phase, translation to a clinical trial.

A new aim was created to improve clinical feasibility of the MR-FUS procedure by reducing the procedure time. We developed a novel shear wave imaging technique using MR-ARFI. The method differs from the relative stiffness approach, by providing a quantitative measure of stiffness over a large area, and by reducing validation time. For example, the validation time for a breast tumor requiring 20 boundary markers was reduced from 30 min with the relative stiffness technique, to ~3 min with the shear wave MR-ARFI technique. While this new technique does not validate each lesion individually, rather the adjacent area, it is fast, measures tissue velocity directly, and could be sufficient for the purpose of pre-surgical validation of MR-FUS markers.

The future directions for the work are to develop a full clinical protocol including imaging and histology, for a pilot study in women scheduled to receive breast conservation surgery.



Toward MR-guided high intensity focused ultrasound for presurgical localization: focused ultrasound lesions in cadaveric breast tissue.
Periodical:Journal of Magnetic Resonance Imaging
Index Medicus: J Magn Reson Imaging
Authors: Bitton RR, Kaye E, Dirbas FM, Daniel BL, Pauly KB
Yr: 2012 Vol: 35 Nbr: 5 Abs: Pg:1089-97