Diffusion Tensor MRI to Distinguish Brain Tumor Recurrence From Radiation Necrosis

The most common major side effect of radiation therapy for the treatment of primary brain
tumors is the necrosis of normal brain tissues (radiation necrosis). Radiation necrosis
typically occurs weeks to months following treatment. The diagnosis is suspected when
patients have new areas of gadolinium enhancement on magnetic resonance imaging (MRI) scans.
Tumor recurrence can also occur within weeks to months following treatment and is represented
by new areas of gadolinium enhancement as well. As the management options for radiation
necrosis and tumor recurrence are significantly different, distinguishing the two is of
critical importance. Conventional MRI, diffusion-weighted MRI (DW-MRI), MR spectroscopy,
SPECT, and PET imaging have all been used to try to make this distinction, but the
sensitivity and specificity of these techniques have not been clinically useful. Most
patients must therefore undergo a risky diagnostic surgical procedure. Diffusion tensor MRI
(DT-MRI) is an imaging technique that provides information regarding both the diffusive
properties of water as well as the directionality of water movement. A modification of
DT-MRI, referred to as two-compartment DT-MRI, appears to be more sensitive than other
variations of MRI for the diagnosis of selected types of brain abnormalities.

Objective: We plan to conduct a feasibility study to determine if two-compartment DT-MRI can
distinguish tumor recurrence from radiation necrosis.

Study Population: Patients aged greater than or equal to twenty one years old with a prior
diagnosis of primary brain tumor and history of radiation treatment who develop new areas of
gadolinium enhancement on conventional MRI scans and who require surgery for diagnostic or
therapeutic purposes will be evaluated for enrollment in this study. Ten patient controls
will also undergo DT-MRI scans for the purpose of obtaining normative data for this
quantitative study.

Design: Patients who meet eligibility criteria will undergo a two compartment DT-MRI scan.
Regions of abnormality will be identified and surgical biopsies will be obtained of these
regions. The radiographic and histologic characteristics of the samples will be correlated.
Volunteers will undergo DT-MRI scans only.

Outcome Measure: The primary outcome measure of this study is the degree of agreement between
the radiographically predicted diagnosis of tumor recurrence or radiation necrosis using
two-compartment DT-MRI and the histologic diagnosis of such. This outcome will be measured as
the proportion of instances in which the two modalities identify a particular lesion as being
tumor recurrence or as being radiation necrosis. The information gathered from this study
will allow for the implementation of a larger study with more patients if there is a high
degree of agreement between the two compartment DT-MRI and histologic diagnoses of surgical
biopsy specimens. The long range goal of the larger study is to radiographically diagnose
tumor recurrence or radiation necrosis with a high enough level of sensitivity and
specificity to avoid a diagnostic surgical procedure with its attendant risks.

- INCLUSION CRITERIA:

Patients must:

Have a histologically confirmed glioma, for which radiation therapy has been previously
administered.

Be able to undergo an MRI scan of the brain.

Have contrast enhancing lesions that are amendable to surgical biopsy and/or resection.

Be appropriate for an operative procedure as determined by a neurosurgeon and
anesthesiologist.

Ten patient controls will be included in this study to foster technical development and for
the acquisition of normative data. Patient controls will have defined unilateral
abnormalities on previously obtained MRI scans, but will not have had radiation treatment.

Be greater than or equal to 21 years of age.

EXCLUSION CRITERIA:

Patients must not:

Have any of the following: aneurysm clip, implanted neural stimulator, implanted cardiac
pacemaker or auto defibrillator, cochlear implant, ocular foreign body or implant (e.g.
metal shavings, retinal clips), or insulin pump as these items would be contra-indications
to undergoing an MRI scan.

Be poor operative candidates from an anesthetic point of view secondary to other major
medical illnesses - the risk of undergoing general anesthesia outweighs the potential
benefit of the clinical information gained from a surgical biopsy/resection.

Have a coagulopathy demonstrated by an abnormal prothrombin time, activated partial
thromboplastin time, or thrombocytopenia (platelet count less that 150,000 platelets/mm3) -
the risk of developing uncontrollable intra-operative bleeding outweighs the potential
benefit of the clinical information gained from a surgical biopsy/resection.

Significant psychiatric impairments which, in the opinion of the investigators, will
interfere with the proper administration or completion of the protocol - self explanatory.

Acute or untreated infections (viral, bacterial or fungal) - patients with active
infections are highly likely to have spread of their infections to the brain as a result of
a biopsy/resection.

Be pregnant at the time of the treatment - Women who are pregnant or nursing are excluded
from this protocol. Therefore, all women of childbearing potential will have a pregnancy
test performed, which must be negative, before proceeding. General anesthesia and surgery
may subject the fetus to unacceptable risks. Also, the NIH does not offer full obstetrical
services in the event that medical care to the mother and/or fetus is required.