Faculty Bibliography

OBJECT: The study was conducted to determine the association between dual-isotope single-photon emission computerized tomography (SPECT) scanning and histopathological findings of tumor recurrence and survival in patients treated with high-dose radiotherapy for glioblastoma multiforme. METHODS: Studies in which SPECT with 201Tl and 99mTc-hexamethypropyleneamine oxime (HMPAO) were used were performed 1 day before reoperation in 47 patients with glioblastoma multiforme who had previously been treated by surgery and high-dose radiotherapy. Maximum uptake of 201Tl in the lesion was expressed as a ratio to that in the contralateral scalp, and uptake of 99mTc-HMPAO was expressed as a ratio to that in the cerebellar cortex. Patients were stratified into groups based on the maximum radioisotope uptake values in their tumor beds. The significance of differences in patient gender, histological characteristics of tissue at reoperation, and SPECT uptake group with respect to 1-year survival was elucidated by using the chi-square statistic. Comparisons of patient ages and time to tumor recurrence as functions of 1-year survival were made using the t-test. Survival data at 1 year were presented according to the Kaplan-Meier method, and the significance of potential differences was evaluated using the log-rank method. The effects of different variables (tumor type, time to recurrence, and SPECT grouping) on long-term survival were evaluated using Cox proportional models that controlled for age and gender. All patients in Group I (201Tl ratio < 2 and 99mTc-HMPAO ratio < 0.5) showed radiation changes in their biopsy specimens: they had an 83.3% 1-year survival rate. Group II patients (201Tl ratio < 2 and 99mTc-HMPAO ratio of > or = 0.5 or 201Tl ratio between 2 and 3.5 regardless of 99mTc-HMPAO ratio) had predominantly infiltrating tumor (66.6%); they had a 29.2% 1-year survival rate. Almost all of the patients in Group III (201Tl ratio > 3.5 and 99mTc-HMPAO ratio > or = 0.5) had solid tumor (88.2%) and they had a 6.7% 1-year survival rate. Histological data were associated with 1-year survival (p < 0.01): however, SPECT grouping was more closely associated with 1-year survival (p < 0.001) and was the only variable significantly associated with long-term survival (p < 0.005). CONCLUSIONS: Dual-isotope SPECT data correlate with histopathological findings made at reoperation and with survival in patients with malignant gliomas after surgical and high-dose radiation therapy.

Inflammatory lesions of the hypophysis include lymphocytic hypophysitis, pituitary abscess, and granulomatous inflammation, with or without specific infections (i.e., sarcoidosis, mycobacteria). These lesions are known to mimic pituitary neoplasms. We report the clinical and pathologic findings in three patients who underwent transsphenoidal resection for presumed pituitary adenoma. Two were women aged 30 years (one with a 5-month history of headache, the other with a 1-year history of menstrual irregularity) and one was a 12-year-old girl with headache, nausea, and diabetes insipidus. Preoperative endocrinologic studies showed increased prolactin in one patient and normal serum thyroid stimulating hormone and prolactin levels in another. By magnetic resonance imaging (MRI), the first case had a 1.2-cm mass with increased signal on T1 and isointensity on T2, ring enhancement after gadolinium, and lateral deviation of the pituitary stalk. The second patient had a 1.1-cm "cystic" mass seen during magnetic resonance imaging with adjacent bony changes seen during computed tomography. In the third, computed tomography showed a hypodense pituitary mass that enlarged during 1-month observation. At surgery, abnormal soft tissue surrounded liquefied material in the anterior pituitary in all cases. Histologic studies showed fragments of intact normal anterior pituitary with preserved vascular and reticulin network and regions of anterior pituitary infiltrated by foamy histiocytes. Other fragments resembled granulation tissue, and some consisted of acellular debris. Histiocytes were immunoreactive for the macrophage marker CD68 and negative for S-100 and CD1a. Ultrastructurally, the normal adenohypophysis was permeated by lipid-laden macrophages. There were no well-formed granulomas or giant cells, hemosiderin, acid-fast bacilli, or fungi. Serial sections and keratin immunostains failed to identify an epithelial cyst lining or keratin among the debris. We propose the term "xanthomatous hypophysitis" for this lesion.

OBJECTIVE: To assess the efficacy of 10% formalin perfusion fixation as a method of rapid fixation to examine the human brain immediately following autopsy. DESIGN: Compare the histology and immunohistochemistry from human brains in which one hemisphere undergoes perfusion fixation using 10% buffered formalin, and the contralateral nonperfused hemisphere undergoes standard 14-day immersion fixation in 8 L of 10% buffered formalin. SETTING: Autopsy material in a general medical-surgical university hospital. PARTICIPANTS: Pathologists, neuropathologists, resident pathologists, and pathology assistants. INTERVENTION: Immediately following brain removal, a single hemisphere was perfused with 1 L 10% buffered formalin over a 15- to 20-minute period. The contralateral nonperfused hemisphere served as a control, undergoing standard immersion fixation for 2 weeks in 10% formalin. The perfusion-fixation hemisphere was immediately available for neuropathologic examination, and histologic sections of the brain were processed immediately with the other necropsy tissue sections. This allows completion of a final autopsy neuropathology report within 3 to 5 days in concert with the systemic section of the report. MAIN OUTCOME MEASURE: Perfusion-fixation brain sections were compared with immersion-fixation brain sections from the same brain. The effects on hematoxylin-eosin, Bielschowsky's silver, and immunohistochemical staining were evaluated by an experienced neuropathologist and a general pathologist with no prior knowledge of the fixation technique. RESULTS: Perfusion fixation revealed equal and occasionally superior histologic sections compared with traditional immersion fixation in terms of (1) technical preparation of section, (2) quality and intensity of staining with both hematoxylin-eosin and silver, and (3) immunoreactivity localization with a variety of immunohistochemical reactions. CONCLUSIONS: Immediate perfusion of the brain is an easily performed fixation technique that yields comparable or superior fixation to prolonged immersion fixation and allows an immediate complete neuropathologic examination and report within 3 to 5 days of performance of the autopsy.

Gangliogliomas are rare tumors of the central nervous system that account for approximately 1% of all brain tumors. Histologically, gangliogliomas are composed of intimately admixed glial and neuronal components, the pathological origins of which remain to be characterized. Clonal analysis through examination of the pattern of the X chromosome inactivation allows one to distinguish monoclonal differentiation of a genetically abnormal progenitor cell from parallel, but independent, clonal expansion of two different cell types during tumorigenesis in biphasic neoplasms, such as gangliogliomas. In the present study, we investigated the clonality of eight gangliogliomas from female patients using both methylation- and transcription-based clonality assays at the androgen receptor locus (HUMARA) on the X chromosome. Among tumors from seven patients who were heterozygous at the HUMARA locus, five were identified as monoclonal with the methylation-based clonality assay, and the results were confirmed by the transcription-based method, whereas two were shown to be polyclonal by the methylation-based clonality assay but monoclonal by transcription-based clonality analysis. We conclude that the predominant cell types in most gangliogliomas are monoclonal in origin and derive from a common precursor cell that subsequently differentiates to form neoplastic glial and neuronal elements.