Comparison of fluorescent in situ hybridization (FISH) and the polymerase chain reaction (PCR) for detection of residual neuroblastoma cells.

Residual neuroblastoma (NB) cells in bone marrow or peripheral progenitor hematopoietic cells harvests may be a source of relapse after autologous transplantation in patients with high stage neuroblastoma. Therefore a sensitive method for detecting minimal residual disease (MRD) in the harvested product is important so that an appropriate purging techniques can be applied and optimized. We report on the detection of NB cells by fluorescence in situ hybridization (FISH) for N-myc amplification and compare the sensitivity of FISH with a semiquantitative polymerase chain reaction (PCR) assay. As a model of MRD we used the neuroblastoma cell line IMR-32 diluted with normal peripheral blood lymphocytes. We were able to detect a single NB cell in 1000 normal mononuclear cells by FISH. The PCR method, using ethidium-bromide-stained gels, required at least 10% NB cells to be present for detection of an amplified band of the N-myc oncogene. Thus, FISH is ten to one hundred times more sensitive in detection of N-myc amplification than a differential PCR and thus it is the method of choice for detection of MRD in NB patients.