A number of recent reports have demonstrated that only CD133-positive cancer cells of glioblastoma multiforme (GBM) have tumor-initiating potential. These findings raise an attractive hypothesis that GBMs can be cured by eradicating CD133-positive cancer stem cells (CSCs), which are a small portion of GBM cells. However, as GBMs are known to possess various genetic alterations, GBMs might harbor heterogeneous CSCs with different genetic alterations. Here, we compared the clinical characteristics of two GBM patient groups divided according to CD133-positive cell ratios. The CD133-low GBMs showed more invasive growth and gene expression profiles characteristic of mesenchymal or proliferative subtypes, whereas the CD133-high GBMs showed features of cortical and well-demarcated tumors and gene expressions typical of proneuronal subtype. Both CD133-positive and CD133-negative cells purified from four out of six GBM patients produced typical GBM tumor masses in NOD-SCID brains, whereas brain mass from CD133-negative cells showed more proliferative and angiogenic features compared to that from CD133-positive cells. Our results suggest, in contrast to previous reports that only CD133-positive cells of GBMs can initiate tumor formation in vivo CD133-negative cells also possess tumor-initiating potential, which is indicative of complexity in the identification of cancer cells for therapeutic targeting. A recent concept in brain tumor biology is that brain tumors arise from cancer stem cells (CSCs) that are CD133 positive (CD133 ( þ ) ). It has been reported that a small number of CD133 ( þ ) glioblastoma multiforme (GBM) cells are able to recapitulate the original tumor in vivo, whereas millions of CD133-negative (CD133 (À) ) cells could not produce brain tumor masses. 1-6 However, accumulating evidence suggests that CD133 (À) GBM cells can also regenerate heterogenous tumors in vivo, 7,8 and generation of the huge and rapidly growing tumors by only CD133 ( þ ) CSCs would be difficult because more than 50% of GBM patients have few CD133 ( þ ) cells. 9 As a majority of neurogenic astrocytes in the adult brain are not recognized by a CD133 antibody, 8 it is likely that CD133 might be newly expressed in GBM CSCs that are derived from CD133 (À) adult neural stem cells (NSCs) or terminally differentiated brain cells, such as astrocytes, neurons, and oligodendrocytes. Given that the gene expression profile is changed when GBM recurs after treatments, 10 it is plausible that new CD133 expression may occur if the characteristics of CSCs are changed or if some CSCs are selected by treatment. Furthermore, the wide-range variation in CD133 ( þ ) cell ratio (0.1-50% in GBM patients) 1-6 also suggests the existence of other GBM CSCs that do not express CD133.Therefore, we hypothesize that there are several kinds of CSCs in the tumor mass of GMB, and these diverse CSCs