2010 年 20 巻 1 号 p. 49-56
It was first pointed out by Inokuchi et al. that ploidy differences in superficially spreading and penetrating gastric cancers may reflect different genetic constitution. Further studies, however, revealed that such a correlation is typically seen only in the undifferentiated-type and that ploidy can alter during tumor progression. DNA ploidy alteration is now considered as an aspect of chromosomal instability. By mapping cytometrically determined DNA ploidy of metaphase cells in individual cancers, we found that (1) DNA-aneuploidy was rare in early cancers and accounted for 70% of advanced cancers, and was almost always accompanied by DNA-diploid cancer cells in the mucosa, suggesting that advanced cancers with aneuploidy may be preceded by DNAdiploid early cancers. (2) The incidence of cycling polyploid cells accompanied in the stemline may be more important than the ploidy mode, itself, for an assessment of chromosomal instability. In DNA-diploid mucosal and extramucosal parts of advanced cancers, the incidences of cycling polyploid cells were as low as that in DNA-diploid early cancers and as high as that in DNA-aneuploid part, respectively. To demonstrate that cycling polypoid cells precede DNA-aneuploidy, we examined the distribution of the cells with duplicated chromosomespecific FISH signals in tissue sections by confocal microscopy. It was found that the cells with duplicated signals were distributed either sporadically or in small clusters. The latter pattern may represent cycling polyploid populations, from which a DNA-aneuploid subclone with growth advantages may arise through further alterations of modal chromosome number. Our studies suggest that the time when DNA-aneuploidy occurs in the tumor and the degree of chromosomal instability (incidence of cycling polyploidy) may be more important clinically than the presence or absence of DNA-aneuploidy.