Dl. Silbergeld et Mr. Chicoine, ISOLATION AND CHARACTERIZATION OF HUMAN-MALIGNANT GLIOMA-CELLS FROM HISTOLOGICALLY NORMAL BRAIN, Journal of neurosurgery, 86(3), 1997, pp. 525-531
Brain invasion prevents complete surgical extirpation of malignant gli
omas; however, invasive cells from distant, histologically normal brai
n previously have not been isolated, cultured, and characterized. To e
valuate invasive human malignant glioma cells, the authors established
cultures from gross tumor and histologically normal brain. Three men
and one woman, with a mean age of 67 years, underwent two frontal and
two temporal lobectomies for tumors, which yielded specimens of both g
ross tumor and histologically normal brain. Each specimen was acquired
a minimum of 4 cm from the gross tumor. The specimens were split: a p
ortion was sent for neuropathological evaluation (three glioblastomas
multiforme and one oligodendroglioma) and a portion was used to establ
ish cell lines. Morphologically, the specimens of gross tumor and hist
ologically normal brain were identical in three of the four cell cultu
re pairs. Histochemical staining characteristics were consistent both
within each pair and when compared with the specimens sent for neuropa
thological evaluation. Cultures demonstrated anchorage-independent gro
wth in soft agarose and neoplastic karyotypes. Growth rates in culture
were greater for histologically normal brain than for gross tumor in
three of the four culture pairs. Although the observed increases in gr
owth rates of histologically normal brain cultures do pot correlate wi
th in vivo behavior, these findings corroborate the previously reporte
d stem cell potential of invasive glioma cells. Using the radial dish
assay, no significant differences in motility between cultures of gros
s tumor and histologically normal brain were found. In summary, tumor
cells were cultured from histologically normal brain acquired from a d
istance greater than 4 cm from the gross tumor, indicating the relativ
e insensitivity of standard histopathological identification of invasi
ve glioma cells (and hence the inadequacy of frozen-section evaluation
of resection margins). Cell lines derived from gross tumor and histol
ogically normal brain were usually histologically identical and demons
trated equivalent motility, but had different growth rates.