ABSTRACT
The prevailing molecular theory of tumors postulates that cancer is a genetic disease caused by mutations in genes belonging to three main families: oncogenes, tumor-suppressor genes, and stability genes. Oncogenes and tumor-suppressor genes encode proteins that regulate cell number in tissues; that is, the balance between cell increase (proliferation) and loss (apoptosis). Stability genes (often referred to as “caretaker” genes) encode proteins that maintain the integrity of the genome through the monitoring and repair of lesions (1). Each family includes an ample (tens) but defi ned number of genes. A tumor derived from a single patient usually contains alterations in multiple genes, combining members of the three families. It is believed that no single gene alteration is capable, alone, to drive the entire process of transformation of a normal cell into a fully malignant cancer cell (1). Cancer malignancy is indeed a complex phenotype characterized by multiple traits, which can be summarized as the ability (i) to autonomously increase in cell number, independently of extracellular
signaling, (ii) to proliferate for an unrestricted number of cell cycles (so-called “replicative immortality”), (iii) to induce the formation of new vessels, and (iv) to trespass normal tissue boundaries. Tissue invasion is a prerequisite for the formation of secondary tumors in distant organs (metastasis) (2). In the last 10 years, experimental proof has been provided to support the old notion that, in the tumor mass, not all the cells are equal, and only a small cell subset is necessary and suffi cient for the generation of the primary tumor and its metastases. These are called “tumor-initiating cells” or “cancer stem cells,” by analogy with normal stem cells that are responsible for regeneration of tissues, either by default (such as in the bone marrow or epidermis) or on demand (such as in the nervous tissue) (3). Cancer stem cells divide, giving rise to two kinds of progeny: cells that replenish the cancer stem cell pool (that is, they self-renew and are endowed with replicative immortality), and cells destined to aberrantly differentiate into the heterogenous and nontumorigenic cell types of the tumor bulk. If unaffected by therapies, cancer stem cells cause tumor relapse, while their “differentiated” progeny, although numerically predominant, should be relatively innocent. The cancer stem cell paradigm predicts that this is the cell that accumulates genetic lesions responsible for cancer, and that this cell must be targeted by therapies in order to cure the patient (4,5).
