RESUMO
BACKGROUND: It has been suggested that the donor tissue cores used in tissue microarrays (TMAs) may not be representative of the whole tissue section. AIM: To validate the use of TMA technology in the study of malignant peripheral nerve sheath tumours (MPNSTs). METHODS: A TMA was constructed containing five independent core biopsy samples of 14 formalin-fixed, paraffin-embedded MPNSTs. The immunohistochemical (IHC) results of the five cores from the same tissue block on TMA were compared with readings from whole sections using two antibodies: anti-Ki-67 and anti-S-100. Digital image analysis was performed to calculate the percentage of positive stain areas. The agreement between IHC results obtained with TMA cores and whole sections was assessed using the kappa statistic. RESULTS: There was good to very good agreement between IHC results for whole and TMA sections from MPNSTs. In relation to S-100, very good agreement (92% agreement; kappa = 0.77) was observed using a minimum of four TMA cores. Staining results for Ki-67 from at least four readable TMA cores were the same as those for the whole section in 86% of cases, with good agreement, using weighted kappa statistics (kappa = 0.63). CONCLUSIONS: This study indicates that the TMA technique can be used in the IHC study of MPNSTs, even with the use of heterogeneous markers such as S-100 protein.