Detecting chromosomal alterations at 1p and 19q by FISH and DNA fragment analysis--a comparative study in human gliomas.

Histological classification of gliomas is important for treatment and as a prognostic predictor, but classification by histology alone can be a challenge. Molecular genetic investigations, in particular the combined loss of the short arm of chromosome 1 and the long arm of chromosome 19 (LOH1p/19q), has become a significant predictor of outcome in oligodendrogliomas. 1p/19q alterations can be investigated by fluorescence in situ hybridization (FISH), but controversies persist in the interpretation ofresults. Another technique is polymerase chain reaction (PCR) analysis using microsatellites as primers and capillary electrophoresis or southern blot as detection method. The objective of the present study was to compare the accuracy, reliability and feasibility of detecting chromosomal changes at 1p/19q with PCR microsatellite analysis and FISH in glial tumors in the clinical laboratory, where often only small formalin-fixed paraffin-embedded samples are available. Commercial DNA and normal cortex were used for comparison. The material comprised 41 glial tumors including 10 oligodendrogliomas (WHO Grades II and III, 5 each), 10 mixed oligoastrocytomas (WHO Grades II and III, 5 each), 10 astrocytomas (WHO Grades II and III, 5 each), and 11 glioblastomas (WHO Grade IV). Our data confirmed a correlation between FISH and LOH fragment analysis in classical oligodendrogliomas and in mixed oligoastrocytomas. Disparity was found among the glioblastomas, where fragment analysis showed 1p/19q loss in three cases, with no changes detected by FISH. The fragment analysis seems reliable and implementable for LOH 1p/19q investigation without patient-related control material.