Amplification of c-myc oncogene by chromogenic and fluorescence in situ hybridization in archival breast cancer tissue array samples.

Fluorescence in situ hybridization (FISH) is currently considered to be the most specific and sensitive method for detection of oncogene amplifications in human tumor samples. However, FISH requires fluorescence microscopy, which is tedious and does not allow histopathologic evaluation of the cells and tissues examined. Here we compared FISH with the newly developed chromogenic in situ hybridization (CISH), which uses peroxidase enzyme for probe detection instead of fluorescent dyes. CISH was found to be highly concordant with FISH in a tissue array series of 177 archival breast cancer samples. This was true both when comparing CISH with single-color and two-color FISH, the latter including the chromosome 8 centromere probe as reference (the kappa coefficients were 0.67 and 0.76, respectively). Clinicopathologic correlations of c-myc amplification as detected by FISH and CISH were generally the same. By both methods, c-myc amplification was significantly associated with high histologic grade, negative progesterone receptor status, DNA aneuploidy, and high S-phase fraction. c-myc amplification was strongly associated with poor distant metastasis-free survival when amplification was detected by CISH (p = 0.0013), but this association was weaker when FISH was used (p = 0.16 for two-color FISH and p = 0.065 for single-color FISH). These data suggest that CISH is at least as sensitive and specific as FISH in the detection of oncogene amplification in human tumor samples. The possibility for concomitant tissue architecture evaluation using an ordinary transmitted light microscope may favor the use of CISH over FISH in oncogene amplification detection in large tumor series, and tissue arrays and, ultimately, in routine clinical diagnostics.