Search

Background: Ischemic stroke remains a major cause of disability with 90% of patients achieving no improvement after 90 days. The primary objective of this study was to evaluate the safety and feasibility of intraparenchymal administration of SB623 cells in chronic stroke. SB623 cells are adult bone-marrow-derived cells transfected with a plasmid encoding the intracellular domain of Notch-1 that have demonstrated improvement in functional testing in animal models of stroke. Methods: Patients with hemiparesis from stable subcortical ischemic stroke and NIHSS > 7 received intracranial injection of 2.5, 5 or 10 million SB623 cells. Clinical outcome measures included NIHSS, MRS, ESS and FMA measured at baseline and repeated at 6 and 12 months. FDG-PET was analyzed at each time point using cluster analysis within 5 regions of interest (contralateral frontal cortex, medial ventral to infarct, contralateral putamen, contralateral sensori-motor cortex, and contralateral thalamus). Results: Five patients enrolled at one academic center were included in this initial FDG-PET analysis. In these 5 patients, there was improvement in NIHSS (p = 0.008) and ESS (p = 0.006), which was most apparent from 0 to 6 months (p = 0.005; p = 0.005). There was also an increase in FDG-PET activity in the contralateral sensori-motor cortex from 0-6 months (p = 0.005), which was related to NIHSS, ESS and FMA (p = 0.043). Analyses of individual patients showed increased FDG uptake contralateral to the injection site in 3/5 patients. There was an increase near the area of infarct in 2/5 patients that was seen at 6 months but not at 12 months. Conclusions: Intraparenchymal injection of SB623 cells in chronic ischemic strokes resulted in increased FDG uptake predominantly in areas contralateral to stem cell injection in some patients. These increases were associated with improved clinical outcome measures suggesting that the PET changes might reflect activation of neural pathways due to stem cell therapy. Further PET studies with appropriate controls are warranted.

BACKGROUND A retrospective study of patients undergoing stereotactic radiosurgery for one to four brain metastases from nonsmall lung cell carcinoma (NSCLC) was performed to document outcomes and risks. METHODS Seventy-seven patients underwent radiosurgery during a 7-year interval; 71 also underwent whole brain radiation therapy. Univariate and multivariate analyses were used to determine significant prognostic factors affecting survival. RESULTS The overall median survival was 10 months after radiosurgery, and 15 months from the diagnosis of brain metastases. Five factors significantly affected survival: extent of systemic disease, presence of a neurologic deficit, size of the intracranial tumor, initial imaging appearance of intratumoral necrosis, and initial resection of the primary tumor of the chest. Median survival time was 26 months in a subgroup of patients with no extracranial metastases, no neurologic deficits, and a small tumor without necrosis. The authors evaluated 91 tumors with imaging. Local tumor control was achieved in 77 lesions (85%) and tumoral radiation necrosis developed in 4 lesions (4.4%). Nineteen new metastatic tumors developed during the observation interval. CONCLUSIONS Stereotactic radiosurgery for NSCLC brain metastases is effective and is associated with few complications. The early detection of brain metastases and treatment with radiosurgery combined with radiation therapy provide the opportunity for extended high quality survival. Cancer 1997; 80:2075-83. © 1997 American Cancer Society.

Alterations in gyral contour made it difficult to identify the motor cortex thought to be near an arteriovenous malformation (AVM) in a 24-year-old man considered for stereotactic radiosurgery. Functional imaging in three modalities was performed preoperatively to compare the reliability of localization using functional magnetic resonance imaging (fMRI) on a conventional scanner with positron emission tomography (PET) and magnetoencephalography (MEG). Similar tasks were used for each imaging modality in an attempt to activate and identify the sensory and motor cortex. Data from all three modalities converged for the sensory task, and fMRI and PET data converged for the motor task. The right hemisphere motor strip was localized adjacent and anterior to the AVM. These data were used in planning the radiosurgery isodose configuration to the AVM in order to reduce the irradiation of motor cortex parenchyma. A postoperative fMRI study was also performed using newer techniques to reduce head motion artifact and to improve signal-to-noise ratio. The data confirmed the conclusions derived from the preoperative evaluations. This study demonstrates how conventional MRI scanners can be used for functional studies of use in surgical planning.