The researchers' specific aims are to test the following hypotheses:

Hypothesis 1: A tumor can be completely ablated by ILT with MRI-guidance.

Hypothesis 2: The MRI-based 3D temperature map of tissue during ILT is predictive of destruction.

Hypothesis 3: The 3D "thermal dose" map that is based on the tissue's temperature over time is more predictive of tissue destruction than the temperature map.

The goal is to evaluate the use of minimally invasive interstitial laser therapy (ILT) in the brain. Our group is in a unique position to offer image-guided ILT because of our expertise and resources here at Brigham & Women's Hospital in the Departments of Neurosurgery and Radiology. The therapy will be monitored and controlled by the use of magnetic resonance imaging (MRI). ILT is a minimally invasive procedure in which the targeted tissue is thermally destroyed in situ in a controlled fashion. The intra-operative MRI provides a way to "see" the treatment. It can be used to treat disease by guiding surgery by providing images of tissue changes during therapy.

In spite of its appeal as a minimally invasive technique, MRI-guided ILT is not commonly practiced in the United States. One reason is that proper clinical implementation of ILT requires an operating room (OR) setting and an MRI scanner - a very rare combination. Our MRI-OR suite includes a sterile procedure room with a 0.5 Tesla vertically "open" magnet. In the past, we have performed MRI-guided ILT procedures in 9 patients. While few in number, this is the most extensive U.S. experience in ILT in the brain.

We have recently created a new image networking and display package for the visualization of 3D information during laser therapy. This provides a view of multiple image planes taken through the tissue volume around the fiber tip from which the light emits. Recent FDA approval for a "diffusing tip" laser and its associated catheter, provides us with an added tool for the procedure. This diffusing technology had been available in Europe for clinical use for years; however, there had been no FDA-approved device in the US until now. The catheter is important because it protects the fiber; it is placed into the tumor first under MRI guidance. Then the laser fiber inserted into the catheter to deliver the light. All equipment used in this protocol is now FDA approved.

Each patient will undergo ILT. The procedure will be performed under anesthesia as per standard procedures. The surgical placement of the laser fiber is a procedure identical to the well-developed and practiced technique of brain biopsy. A hole approximately 1 cm in diameter will be drilled in the skull through which the laser fiber will be placed under image guidance to confirm the actual progress during the advance of the fiber. We will deliver energy at a rate and distribution of 1-12 watts/cm for exposures less than 20 minutes. After the laser has been turned off, and the tissue cooled, MRI will show the region of ablation. As needed, the laser fiber will be moved/re-located to assure that the total target has been ablated. After the treatment is complete, the fiber is withdrawn, final images are acquired and the surgical site is closed and dressed. On the day after the procedure, the patient will undergo a 24 hour follow-up MRI exam. There will be post-operative care as with any neurosurgical patient.

The following continuous variables will be measured in this study:

• The pre-operative tumor volume (VO) in cc
• The post-operative ablated volume (V1) in cc
• The intra-operative critical temperature volume (VT) in cc
• The intra-operative critical dose volume (VD) in cc

The following statistical hypothesis tests will be conducted.

Statistical Hypothesis 1. A tumor can be completely ablated by ILT with MRI-guidance.

We propose that the difference between the mean pre-op tumor volumes and the post-op ablated volumes (VO and V1, respectively) is zero. Residual tumor is defined as (V0-V1). This will be determined by calculating the mean of the values of the proportion of residual tumor, defined as (V0-V1)/ V0. Use of the proportion normalizes the data for different sized tumors.

Statistical Hypothesis 2. The MRI-based 3-D temperature map of the tissue during ILT is predictive of destruction.

We propose that the difference between the mean post-op ablated volumes and the intra-operative critical temperature volumes (VT and V1, respectively) is zero. This will be determined by calculating the mean of the values of the proportion of the difference between them, defined as (VT-V1)/VT.

Statistical Hypothesis 3. The thermal dose map is predictive of tissue destruction.

We propose that the difference between the mean post-op ablated volumes and the intra-operative critical dose volumes (VD and V1, respectively) is zero. This will be determined by calculating the mean of the values of the proportion of the difference between them, defined as (VD-V1 /VD).

Also, data will be collected through Neurological Examinations.

Inclusion Criteria:

• Male or female
• Age 18+
• Surgically difficult to access tumors including intracerebral metastases

Exclusion Criteria:

• Patients unwilling or unable to give written consent
• Patients at risk for cardiac ischemia
• Patients who cannot physically fit in the MRI scanner in the MRI OR
• Patients with contraindications to MRI imaging such as pacemakers, non-compatible aneurysm clips, shrapnel, and other internal ferromagnetic objects
• Patients with coagulopathies, severe medical problems, cardiac arrhythmias or abnormal BUN