Faculty Bibliography

PURPOSE: To describe the MR imaging findings in a pilot study evaluating gene therapy for treatment of patients with recurrent glioblastoma. METHODS: Serial MR examinations were evaluated retrospectively in patients treated with gene therapy that included a retroviral vector containing the herpes simplex virus thymidine kinase gene and intravenous ganciclovir. Images were obtained after tumor resection and after each cycle of treatment, at approximately 40-day intervals. The volume of enhancing tissue was measured on serial MR images. RESULTS: Eleven patients with recurrent glioblastoma were entered into the clinical trial of gene therapy and seven patients completed at least two cycles of treatment. Of these seven, three patients had an early (between 40 and 80 days) increase in the volume of enhancing tissue followed by a decrease or plateau in enhancing tissue volume. A fourth patient had a stable volume of enhancing tissue for 132 days. The remaining three patients had continuous increases in volume of enhancement on all subsequent MR examinations. CONCLUSION: Although animal data show striking tumor regression in response to similar gene therapy, only limited regression was observed among the seven patients we studied. The transient increases in enhancement seen in three of seven patients might reflect an inflammatory response to local injection of the viral vector

RadNotes is a novel software development tool that enables physicians to develop teaching materials incorporating text and images in an intelligent, highly usable format. Projects undertaken in the RadNotes environment require neither programming expertise nor the assistance of a software engineer. The first of these projects, Thoracic Imaging, integrates image teaching files, concise disease and topic summaries, references, and flash card quizzes into a single program designed to provide an overview of chest radiology. RadNotes is intended to support the academic goals of teaching radiologists by enabling authors to create, edit, and electronically distribute image-oriented presentations. RadNotes also supports the educational goals of physicians who wish to quickly review selected imaging topics, as well as to develop a visual vocabulary of corresponding radiologic anatomy and pathologic conditions. Although Thoracic Imaging was developed with the aim of introducing chest radiology to residents, RadNotes can be used to develop tutorials and image-based tests for all levels; create corresponding World Wide Web sites; and organize notes, images, and references for individual use

RATIONALE AND OBJECTIVES. To investigate the mechanism of anaphylactoid reactions to contrast media, in vitro histamine release induced by magnetic resonance imaging, and iodinated contrast agents was examined in a dog mastocytoma cell line. METHODS. Two gadolinium (Gd)-based magnetic resonance contrast agents, Gd diethylenetriamine pentaacetic acid (Gd-DTPA), dimeglumine, and Gd-bismorpholide, and two iodinated contrast agents, diatrizoate meglumine and iohexol, were incubated with histamine-containing canine mastocytoma cells. Release of histamine into the supernatant was determined at various contrast-medium concentrations after incubation at 37 degrees C for 30 minutes. RESULTS. Iodinated and Gd-based contrast agents caused release of histamine from mastocytoma cells at similar concentrations (50-150 mM). Mannitol, an osmotic stimulus, caused release of histamine only at concentrations greater than 1,000 mM. CONCLUSIONS. Histamine release from canine mastocytoma cells does not appear to be solely due to osmotic effects, but results from direct stimulation by contrast media. For all agents examined, the concentration at which in vitro histamine release occurs far exceeds the serum contrast media concentration expected in routine clinical application. Direct release of histamine from mast cells does not completely explain the pathogenesis of idiosyncratic anaphylactoid responses to contrast media

To evaluate the immunogenic potential of gadolinium-based magnetic resonance imaging (MRI) contrast agents, Sprague-Dawley rats were sensitized with gadolinium diethylenetriamine pentaacetic acid (Gd-DTPA) dimeglumine and with Gd-DTPA covalently linked to either human serum albumin, dextran, or polylysine. IgG antibodies directed against Gd-DTPA were detected in immune sera by an enzyme-linked immunosorbent assay (ELISA), and were confirmed by competitive inhibition of antibody binding using free Gd-DTPA dimeglumine. Antiserum induced by immunization with human serum albumin-(Gd-DTPA) was characterized by a monophasic competition curve with 50% inhibition (IC50) = 5.5 x 10(-4) M when Gd-DTPA dimeglumine was used as both the well-coating and the displacing agent in a competition ELISA. Antiserum induced by Gd-DTPA dimeglumine alone was characterized by a biphasic competition curve with IC50 = 6.5 x 10(-7) M and 7.9 x 10(-4) M. Antisera obtained after exposure to either dextran-(Gd-DTPA) or polylysine-(Gd-DTPA) were of insufficient titer for characterization. The detection of antibodies specific for Gd-DTPA suggests in vivo protein binding with formation of hapten-carrier conjugates. This hypothesis is supported by increased relaxivity values observed when Gd-DTPA dimeglumine is incubated in serum rather than in water. Gd-DTPA dimeglumine and albumin-(Gd-DTPA) are immunogenic in rats under idealized experimental conditions. Additional studies will be necessary to determine the potential for immunologic response in humans to gadolinium chelates under conditions of exposure inherent in clinical use

Several recent studies indicate that the von Recklinghausen neurofibromatosis (NF1) gene is located near the centromere of chromosome 17 in some families. However, variable expressivity and a very high mutation rate suggest that defects at several different loci could result in phenotypes categorized as NF1. In order to assess this possibility and to map the NF1 gene more precisely, we have used two polymorphic DNA markers from chromosome 17 to screen several pedigrees for linkage to NF1. We ascertained a large Caucasian pedigree (33 individuals sampled, 17 NF1 affected) as well as eight smaller pedigrees and nuclear families (50 individuals sampled, 30 NF1 affected). Here, we report strong evidence of linkage of NF1 to the centromeric marker D17Z1 (maximum lod = 4.42) and a weaker suggestion of linkage to the ERBA1 oncogene (maximum lod = 0.57), both at a recombination fraction of zero. Since obligate cross-overs with NF1 were not observed for either marker in any of the informative families tested, the possibility of NF1 locus heterogeneity is not supported