✓The authors present three patients who, after excision and irradiation of their brain tumors, were treated with BCNU for recurrence. All three patients responded well and now are without evidence of tumor, 37, 30, and 36 months after BCNU was stopped. Although these patients represent only a small fraction of those treated with BCNU, they indicate the potential role of chemotherapy in the management of glial tumors.
John Calogero, David C. Crafts, Charles B. Wilson, Edwin B. Boldrey, Alan Rosenberg and K. Jean Enot
Theodore W. Eller, Lawrence P. Bernstein, Richard S. Rosenberg and David G. McLone
✓ A case of congenital tethered cervical spinal cord is presented in a young adult. Metrizamide computerized tomography was the most useful imaging technique for identifying the tethered spinal cord. Intraoperative somatosensory evoked potentials correlated well with clinical improvement following surgery.
Stephen C. Saris, Steven A. Rosenberg, Robert B. Friedman, Joshua T. Rubin, David Barba and Edward H. Oldfield
✓ Recombinant interleukin-2 (rIL-2) is an immunotherapeutic agent with efficacy against certain advanced cancers. The penetration of rIL-2 across the blood-cerebrospinal fluid (CSF) barrier was studied in 12 cancer patients who had no evidence of tumor involvement of the central nervous system. At different times during treatment with intravenous rIL-2, CSF was withdrawn either continuously for 8 to 26 hours via a lumbar subarachnoid catheter (in eight patients) or by a single lumbar puncture (in four). Bioassay showed the appearance of rIL-2 in lumbar CSF 4 to 6 hours after the first intravenous dose, a rise over 2 to 4 hours to a plateau of 3 to 9 U/ml, and clearance to less than 0.1 U/ml by 10 hours after the last dose. An abnormally elevated CSF albumin level in two of the twelve patients indicated alteration of the blood-brain barrier. There were no abnormalities in the CSF glucose level or white blood cell count.
The CSF pharmacokinetics contrast with the rapid elimination of rIL-2 from plasma and demonstrate significant blood-CSF barrier penetration. These data support the possibility of achieving CSF levels of rIL-2 that are adequate to maintain activity of lymphokine-activated killer cells after parenteral administration, and argue for rIL-2-associated disruption of the human blood-brain barrier in some patients.
David Barba, Stephen C. Saris, Carol Holder, Steven A. Rosenberg and Edward H. Oldfield
✓ Adoptive immunotherapy using lymphokine-activated killer (LAK) cells and interleukin-2 (IL-2) offers the possibility of a new treatment for patients with malignant glial tumors. In a clinical trial, the effectiveness of a 5-day treatment cycle of direct intratumoral administration of both LAK cells and IL-2 via a reservoir/catheter system in patients with recurrent malignant gliomas was studied. Ten patients were entered into the study, nine of whom were treated with 15 cycles of LAK cells (0.9 to 21.0 × 109 cells) and IL-2 (49 to 450 × 103 U/kg). The 10th patient in the study was not treated because of the onset of severe neurological deficits prior to beginning immunotherapy. Of the nine patients treated, one had a partial tumor response to immunotherapy as documented by computerized tomography. Neurological side effects occurred in all patients undergoing treatment and were related to increases in cerebral edema that appeared to be mediated by the immunotherapy. This report demonstrates the present limitations of regional adoptive immunotherapy with LAK cells and IL-2 in the treatment of human glial tumors.
Stephen C. Saris, Nicholas J. Patronas, Steven A. Rosenberg, Joseph T. Alexander, Joseph Frank, Douglas J. Schwartzentruber, Joshua T. Rubin, David Barba and Edward H. Oldfield
✓ Parenteral treatment with interleukin-2 (IL-2) is effective against certain advanced cancers outside the central nervous system. Prior to commencement of Phase II trials in patients with brain tumors, the neurological and neuroradiological features of 10 patients treated with intravenous administration of repeated doses of IL-2 were studied. Three patients had malignant gliomas, and seven patients had extracranial cancer without evidence of intracranial metastasis. All were treated with intravenous doses of 105 U/kg three times daily for up to 5 days. The patients with gliomas received cranial computerized axial tomography (CT) scans before IL-2 therapy was initiated and during the later stages of treatment. The patients with extracranial cancer under-went T2-weighted magnetic resonance (MR) imaging before and later during therapy.
After two to 11 doses of IL-2, the patients with gliomas had marked neurological deterioration that was associated with a mild to marked increase in peritumoral edema and mass effect visible on CT scans. With cessation of treatment and appropriate supportive care, all returned to their pretreatment state. The patients with extracranial cancer were either neurologically unchanged or underwent minor transient changes in mental status (lethargy and confusion). In these patients, the MR signal intensity was quantified and compared in eight anatomic regions of interest. In six of the seven patients, there were increases in gray and white matter signal intensity consistent with increased cerebral water content. The percentage changes (means ± standard error of the means) were 12.6% ± 7.3% in the gray matter and 17.0% ± 6.2% in the white matter.
This study demonstrates that treatment with a high parenteral dose of IL-2 is not tolerated by patients with gliomas due to increased cerebral edema. In patients with extracranial cancer but no brain disease, parenteral IL-2 induces an increase in the cerebral water content of both gray and white matter.
David N. Louis, E. P. Richardson Jr., G. Richard Dickersin, Debra A. Petrucci, Andrew E. Rosenberg and Robert G. Ojemann
✓ A case of primary intracranial leiomyosarcoma is presented, with clinical, radiological, light microscopic, immunohistochemical, and ultrastructural data. The histogenesis is discussed and the literature on smoothmuscle tumors of the central nervous system is reviewed.
David M. Frim, M. Priscilla Short, William S. Rosenberg, Joseph Simpson, Xandra O. Breakefield and Ole Isacson
✓ Neurotrophic factors, such as nerve growth factor (NGF), in addition to their role in neuronal development, have protective effects on neuronal survival. Intracerebral implantation of cells genetically altered to secrete high levels of NGF is also found to promote neuronal survival in experimental lesioning models of the brain. The range of activity for such biological delivery systems has not yet been well described either spatially or temporally. Therefore, the authors chose to study the local and distant protective effects of an NGF-secreting rat fibroblast cell line implanted in an excitotoxic lesion model of Huntington's disease. They found that preimplantion of NGF-secreting fibroblasts placed within the corpus callosum reduced the maximum crosssectional area of a subsequent excitotoxic lesion in the ipsilateral striatum by 80% when compared to the effects of a non-NGF-secreting fibroblast graft, and by 83% when compared to excitotoxic lesions in ungrafted animals (p < 0.003). However, NGF-secreting cells placed in the contralateral corpus callosum failed to affect striatal lesion size significantly when compared to contralateral or ipsilateral non-NGF-secreting cell implants. Of note, fibroblasts were clearly visible within the graft site at 7 and 18 days after implantation; however, few cells within the grafts stained positively for NGF peptide or for the messenger ribonucleic acid (mRNA) encoding the transfected NGF gene-construct at either time point. These results show that biological delivery systems for NGF appear to have a profound but local effect on neuronal excitotoxicity, which will necessitate careful neurosurgical placement for maximum effect. Furthermore, the ability of this genetically altered cell line to synthesize NGF mRNA and peptide appears to decrease spontaneously in vivo, a characteristic that will need to be addressed before this method of biological delivery can be utilized as a treatment for chronic degenerative diseases.
Ryan G. Chiu, Blake E. Murphy, David M. Rosenberg, Amy Q. Zhu and Ankit I. Mehta
Much of the current discourse surrounding healthcare reform in the United States revolves around the role of the profit motive in medical care. However, there currently exists a paucity of literature evaluating the effect of for-profit hospital ownership status on neurological and neurosurgical care. The purpose of this study was to compare inpatient mortality, operation rates, length of stay, and hospital charges between private nonprofit and for-profit hospitals in the treatment of intracranial hemorrhage.
This retrospective cohort study utilized data from the National Inpatient Sample (NIS) database. Primary outcomes, including all-cause inpatient mortality, operative status, patient disposition, hospital length of stay, total hospital charges, and per-day hospital charges, were assessed for patients discharged with a primary diagnosis of intracranial (epidural, subdural, subarachnoid, or intraparenchymal) hemorrhage, while controlling for baseline demographics, comorbidities, and interhospital differences via propensity score matching. Subgroup analyses by hemorrhage type were then performed, using the same methodology.
Of 155,977 unique hospital discharges included in this study, 133,518 originated from private nonprofit hospitals while the remaining 22,459 were from for-profit hospitals. After propensity score matching, mortality rates were higher in for-profit centers, at 14.50%, compared with 13.31% at nonprofit hospitals (RR 1.09, 95% CI 1.00–1.18; p = 0.040). Surgical operation rates were also similar (25.38% vs 24.42%; RR 0.96, 95% CI 0.91–1.02; p = 0.181). Of note, nonprofit hospitals appeared to be more intensive, with intracranial pressure monitor placement occurring in 2.13% of patients compared with 1.47% in for-profit centers (RR 0.69, 95% CI 0.54–0.88; p < 0.001). Discharge disposition was also similar, except for higher rates of absconding at for-profit hospitals (RR 1.59, 95% CI 1.12–2.27; p = 0.018). Length of stay was greater among for-profit hospitals (mean ± SD: 7.46 ± 11.91 vs 6.50 ± 8.74 days, p < 0.001), as were total hospital charges ($141,141.40 ± $218,364.40 vs $84,863.54 ± $136,874.71 [USD], p < 0.001). These findings remained similar even after segregating patients by subgroup analysis by hemorrhage type.
For-profit hospitals are associated with higher inpatient mortality, lengths of stay, and hospital charges compared with their nonprofit counterparts.