High-throughput molecular profiling of high-grade astrocytomas: the utility of fluorescence in situ hybridization on tissue microarrays (TMA-FISH).

Due to recent biological and technical advances, the list of potentially useful candidate genes is rapidly expanding in the study of brain tumors. However, traditional methods of screening individual genes in individual samples are slow and tedious, often with consumption of precious resources after only a few experiments. This study evaluates the feasibility of high-throughput molecular analysis using fluorescence in situ hybridization (FISH) on glioma tissue microarrays (TMA). A single microarray paraffin block was constructed using 65 WHO grade III and IV astrocytomas, sampled in duplicate with 0.6-mm-diameter punch cores. FISH was used to detect common alterations, such as EGFR amplification, chromosome 7, 9, and 10 aneusomies and deletions of 1p, 19q, PTEN, DMBT1, and p16. Of 585 hybridization sets, 508 (87%) yielded interpretable data, with hybridization failure in 33 (5.5%) and dislodged tissue in 44 sets (7.5%), respectively. Glioblastomas harbored significantly more alterations than anaplastic astrocytomas, with the overall frequencies of alterations similar to those reported using other techniques. The overall concordance rate between paired tumor core samples was 93%. We conclude that TMA-FISH is an efficient and reliable method for detecting molecular alterations in high-grade astrocytomas.