You are looking at 1 - 6 of 6 items for

  • By Author: Oldfield, Edward H. x
  • By Author: Ram, Zvi x
Clear All
Restricted access

Zvi Ram, Thomas H. Shawker, Mary H. Bradford, John L. Doppman and Edward H. Oldfield

✓ Microadenomas of the pituitary vary in size, particularly those related to Cushing's disease. They are often not visualized on preoperative magnetic resonance (MR) imaging and may be difficult to find during surgical exploration of the pituitary. To enhance intraoperative localization of pituitary adenomas, we assessed the feasibility of using ultrasound to detect and localize pituitary tumors. Intraoperative ultrasound (IS) in the axial and sagittal planes was performed with an Intrascan system using a 12-MHz mechanically oscillating, end-firing transducer. Interpretation of the scans was performed by two individuals, who were usually blinded to the results of preoperative MR imaging or petrosal sinus sampling.

Twenty-eight patients were examined. Eighteen of these patients had microadenomas (1.5–7 mm), all with Cushing's disease; nine had macroadenomas (10–20 mm), three of which were adrenocorticotropic hormone—secreting, three growth hormone—secreting, two thyroid-stimulating hormone—secreting, and one nonfunctioning; and one patient had an intrasellar craniopharyngioma. Normal sellar and parasellar structures, such as intrapituitary cysts, the intracavernous carotid arteries, and the diaphragma sella were easily visualized. Twenty-three of the 28 tumors, including 13 of the 18 microadenomas, were detected on IS (82% sensitivity). Tumors were seen as hyperechoic masses in 19 patients, mixed echogenicity in three, and isoechoic in one. In most macroadenomas IS allowed visualization of the interface between the tumor and the normal pituitary gland. These results indicate the potential of IS to aid the intraoperative localization and definition of pituitary tumors.

Restricted access

John J. Viola, Zvi Ram, Stuart Walbridge, Eric M. Oshiro, Bruce Trapnell, Jung-Hwa Tao-Cheng and Edward H. Oldfield

✓ Among the appealing features of adenoviruses as vectors for transfer of genes into the central nervous system (CNS) are that they are not neurotoxic, they can accommodate the insertion of several large genes, they are not associated with the hazards of insertional mutagenesis, and they can be concentrated to a high-titer preparation. The authors evaluated the feasibility of using adenovirally mediated gene transfer into cultured human glioma cells and in rat models of solid brain tumors and meningeal cancer. Replication-deficient adenoviral vector particles carrying a nuclear-localizing lacZ gene were injected into established 9L cerebral gliomas in Fischer rats. In addition, the adenoviral vector was injected into the subarachnoid space, either simultaneously with intrathecal tumor inoculation or after establishing leptomeningeal cancer. The brains and spinal cords were removed at various intervals for histochemical evaluation for β-galactosidase activity using X-Gal staining. Additional rats received a stereotactic intracerebral injection of the vector into normal brain. No clinical abnormalities were observed in the injected rats. Injection of the adenoviral vector into normal brain resulted in diffuse transduction of astrocytes, microglia, neurons, and endothelial cells at the injection site. Injection of a high-concentration vector preparation into cerebral gliomas resulted in effective tumor transduction. Intrathecal injection of the vector in rats with meningeal cancer resulted in transduction of the infiltrating tumor in the subarachnoid space when injections were given simultaneously with, or 7 days after, tumor inoculation. Transduction rates of both solid and leptomeningeal tumors correlated with the number of injected particles. These results suggest that adenoviral vectors can efficiently transduce solid brain tumors and that the vectors survive in the cerebrospinal fluid for a sufficient period of time to allow leptomeningeal tumor transduction. Adenoviral vector should be evaluated for its potential use in therapeutic gene transfer approaches in malignancies of the CNS.

Restricted access

Zvi Ram, Stuart Walbridge, Thomas Shawker, Kenneth W. Culver, R. Michael Blaese and Edward H. Oldfield

✓ Eradication of malignant brain tumors by in situ intratumoral, retrovirally mediated transfer of the herpes simplex virus thymidine kinase (HSVtk) gene, which sensitizes the tumor cells to ganciclovir, has recently been demonstrated in animal models. The observation that tumors studied in vitro and in animals can be completely eliminated despite only partial transduction of the tumor suggests a bystander mechanism that affects nontransduced tumor cells. Such a bystander effect is not completely understood and may represent a combination of several factors that lead to tumor eradication. Endothelial cells of the tumor blood vessels were shown to occasionally integrate the retroviral vector and thus become sensitized to ganciclovir. In the presence of vector-producer cells, which continuously release infectious viral particles, diffuse multifocal hemorrhages occurred during ganciclovir administration. When the tumor was composed of cells that had been transduced with the thymidine kinase gene before inoculation, no infectious viral particles were present within the tumor, no transduction of endothelial cells occurred, and no hemorrhages were observed during ganciclovir therapy. These observations suggest that tumor regression may be due, in part, to destruction of in vivo HSVtk-transduced endothelial cells after exposure to ganciclovir, resulting in tumor ischemia as one possible bystander mechanism.

The authors investigated this hypothesis using the subcutaneous 9L gliosarcoma tumor model in Fischer rats. The tumors were evaluated with Doppler color-flow and ultrasound imaging during the various phases of the study. Twenty rats received intratumoral injections of HSVtk retroviral vector-producer cells (6 × 107 cells/ml) 21 days after bilateral flank tumor inoculation. Ten rats were subsequently treated with intraperitoneal ganciclovir (15 mg/kg/ml twice a day) for 14 days starting on Day 7 after producer cell injection; 10 control rats received intraperitoneal saline injections (1 ml twice a day) instead of ganciclovir. Ultrasound and flow images were obtained before cell injection, before and during ganciclovir or saline administration, and after cessation of treatment. The number, location, and ultrasonographic appearance of tumor vessels and the tumor volumes were recorded.

The number of blood vessels in the tumors increased over time in both groups before treatment. Intratumoral cell injection without ganciclovir administration did not influence tumor growth or intratumoral vasculature. However, tumor vasculature decreased after initiation of ganciclovir therapy in the HSVtk-transduced tumors (p < 0.05). Early patchy or diffuse necrotic changes associated with ultrasonographic evidence of scattered intratumoral hemorrhage occurred in tumors treated with ganciclovir. Reduction of the tumor blood supply may be an important feature of HSVtk transduction-mediated tumor regression and may, at least partially, account for the degree of tumor destruction that occurs despite the lack of transduction of all tumor cells.

Restricted access

Zvi Ram, Stuart Walbridge, John D. Heiss, Kenneth W. Culver, R. Michael Blaese and Edward H. Oldfield

✓ The authors have recently shown the feasibility of eradicating brain tumors using in vivo retroviral-mediated transduction of tumors with the herpes simplex thymidine kinase (HStk) gene and ganciclovir therapy. However, thymidine kinase-transduced subcutaneous tumors in immunocompromised (athymic) mice were less responsive to this therapy than in immunocompetent animals, suggesting a role of the immune system in the process of tumor eradication. Broad suppression of humoral and cell-mediated immunity is found in patients with malignant gliomas. Interleukin-2 (IL-2) production and IL-2 receptor expression are decreased in glioma patients. These findings and the proposed association between lymphocytic infiltration of brain tumors and survival suggest that immune response modifiers may be useful in treating glioma patients.

To evaluate the role of local cytokine expression by tumor cells, alone or combined with HStk gene transfer and ganciclovir therapy, the authors investigated the efficacy of tumor (9L gliosarcoma) eradication in Fischer rats by in vitro and in vivo tumor transduction with the IL-2 gene alone or with a combined vector carrying both the HStk and IL-2 genes. Tumors injected with HStk vector-producer cells alone, with or without ganciclovir, and rats inoculated in the brain and subcutaneously with 9L cells that had previously been transduced in vitro served as controls. Murine vector-producer cells (3 × 106/50 µl) were injected into the brain tumors 7 days after tumor inoculation. Ganciclovir (15 mg/kg) was administered intraperitoneally twice daily for 10 days to animals that received HStk with or without IL-2 vector-producer cells, starting 5 days after producer-cell injection. The experiment was repeated with continuous daily treatment of all rats with oral dexamethasone (0.5 mg/kg). Rats were sacrificed 21 days after tumor inoculation, and the brains were removed for histological and immunohistochemical analysis for IL-2. Within each experimental group, tumors were found in a similar proportion in the dexamethasone-treated and untreated rats. Large brain tumors developed in all 10 rats that had been inoculated with 9L cells which had been pretransduced in vitro with the IL-2 gene, whereas only three of eight rats receiving subcutaneous inoculation of similar cells developed palpable tumors. No enhancement of tumor eradication was observed by adding the IL-2 gene in the HStk vector construct compared to the use of the vector with HStk alone. Lymphocytic infiltration was absent in all dexamethasone-treated rats but was observed in all treatment groups not receiving steroids. The degree of lymphocytic infiltration was not enhanced by intratumoral injection of IL-2 or IL-2/HStk vector-producer cells.

The findings suggest a limited role, if any, for immune enhancement by transduction with IL-2 to eradicate brain tumors, either used alone or in combination with HStk.

Restricted access

Zvi Ram, Lynnette K. Nieman, Gordon B. Cutler Jr., George P. Chrousos, John L. Doppman and Edward H. Oldfield

✓ The potential efficacy of early repeat transsphenoidal surgery for persistent Cushing's disease has not previously been examined. On 222 patients with no prior pituitary treatment and a preoperative diagnosis of Cushing's disease, 29 (13%) remained hypercortisolemic after an initial transsphenoidal pituitary exploration. Seventeen of these 29 patients underwent further surgery 7 to 46 days after the initial transsphenoidal approach in order to completely excise suspected residual tumor. Patients were followed for 4 to 84 months (mean ± standard deviation, 34 ± 25 months) to document sustained remission or recurrence of Cushing's disease (a urine free cortisol level > 90 µg/day was considered evidence of recurrence).

Of the 17 patients with repeat surgery, 12 (71%) had resolution of hypercortisolism (morning plasma cortisol level < 5 µg/dl); however, in three of these 12, hypercortisolism recurred 5, 12, and 24 months later. In 14 patients a lesion that appeared to be a tumor was identified during the initial procedure or on histological examination. Of these, 12 had immediate resolution of hypercortisolism and nine are still in remission. Three patients, in whom no adenoma could be identified during the initial surgery or an examination of the partial hypophysectomy specimen from the initial surgery, had persistent Cushing's syndrome after the second operation. Seven (41%) of the 17 patients developed hypopituitarism requiring treatment with thyroid hormone, gonadal steroid, or vasopressin replacement.

The low incidence of identification of an adenoma on computerized tomography or magnetic resonance images (three of 17 patients), the failure to find a corticotrophic adenoma during the initial surgery (10 of 17 patients), and the failure of these 17 patients to respond to the initial transsphenoidal surgery suggest that they may comprise a subset of patients who are more difficult to treat successfully with surgery than most patients with Cushing's disease. Despite that, early reoperation induced immediate remission of hypercortisolism in 71% of cases, but did so at the expense of a high risk of hypopituitarism. However, since the alternative treatments (such as radiation therapy, long-term drug therapy, or bilateral adrenalectomy) also have potential adverse effects, early reoperation deserves consideration for the management of persistent Cushing's disease, especially when an adrenocorticotrophic hormone-secreting adenoma was partially excised during the first surgery.

Restricted access

Zvi Ram, Kenneth W. Culver, Stuart Walbridge, Joseph A. Frank, R. Michael Blaese and Edward H. Oldfield

✓ Retroviral-mediated transfer of the herpes simplex virus thymidine kinase (HSVtk) gene into malignant tumors confers drug susceptibility to the antiviral drug ganciclovir. The authors have recently shown that in situ transduction of the rat 9L brain tumor following HSVtk-producer cell implantation led to tumor regression after ganciclovir administration in treated rats. A wide spectrum of potential adverse effects may, however, be associated with the application of this approach to treat brain tumors, including dissemination of the retroviral vector to nontumoral tissues within or outside the central nervous system, proliferation of the injected murine vector-producer cells at the injection site, immune-mediated responses to the implantation of xenogeneic cells, and damage to the brain from toxic by-products of the HSVtk-ganciclovir interaction. These possibilities were investigated using intracerebral and systemic injections of retroviral vector-producer cells carrying the HSVtk or the lacZ gene in mice, rats, and nonhuman primates.

Using the lacZ gene as a reporter gene, no evidence of β-galactosidase activity consistent with vector transduction was detected in any major body organ in the treated mice or rats. Similarly, the HSVtk gene transfer did not result in toxicity, with or without ganciclovir administration. In studies using rat and monkey models, no proliferation of the vector-producer cells occurred after intracerebral injection. Vector-producer cell survival was limited to 7 to 14 days. High-dose steroid therapy did not appear to extend the survival of these xenogeneic cells in rats. No significant inflammatory response was observed in the meninges or brain parenchyma. Endothelial cells were occasionally transduced in brain capillaries adjacent to the injected site of the vector-producer cells. Injection of producer cells into brain tissue elicited mild edema and reactive gliosis surrounding the injection site, which were probably the cause of a transient toxic response arising 4 to 5 days following injection of the producer cells; short-term administration of dexamethasone eliminated that response. No neurological deficits were observed in the rats or primates treated with the HSVtk vector-producer cells, with or without ganciclovir. In primates injected with producer cells, magnetic resonance imaging before, during, and after ganciclovir administration showed minimal and localized breakdown of the blood-brain barrier without significant edema or mass effect. Similarly, histological examination of the monkeys' brains showed no damage to neurons, astroglia, or myelin. Long-term clinical (> 9 months) and radiological (3 months) assessment of the primates has revealed no evidence of toxicity. The results of these studies indicate that intratumoral implantation of HSVtk-producer cells can be attempted for the treatment of brain tumors, without anticipating significant adverse toxicity to normal brain or remote proliferating tissues.