Tumor-Initiating Cells: An Influential Paradigm for Xenograft Research

Xenograft-associated brain tumor studies have been performed for as long as neuro-oncology has been a definable field of cancer research. The earliest studies involved heterotransplantation of human brain tumor cells into immunocompetent hosts 1 and subsequently the still-popular athymic nude mouse hosts. 2 Identifying rodent hosts that maximize our ability to study a variety of brain tumor subtypes in vivo continues to see substantial research activity, as indicated by the study by Yu et al 4 published in the current issue of Neuro-Oncology. In the associated investigation, Rag2 mice, which lack B and T cell function, were used as hosts for engrafting and propagating an ependymoma— a central nervous system tumor subtype whose degree of malignancy is generally insufficient for successful propagation using the conventional athymic mouse model. The mouse host used in the current study, although significant, does not represent the major subject of this editorial. Rather, this editorial focuses on primary tumor and derivative xenograft cancer stem cell, or tumor-initiating cell, composition. In 2004, results presented by Singh et al 3 supported the importance of the CD133+ fraction of patient brain tumors in tumorigenicity in NOD-SCID mice. The pervasive influence of this report has been evident ever since, and the characterization of brain tumors, especially gliomas, and corresponding derivatives for CD133 composition has been a consistent feature of published studies over the past 6 years. In fact, as of April 5, 2010, the manuscript by