Within the past 10-15 years our knowledge about cancer and how cancer cells might originate has changed dramatically. It is now generally believed that a tumor has its origin in cancer stem cells (CSCs), which originated either from transformed tissue stem cells or transformed progenitor cells that have regained self-renewal activity. CSCs share several characteristics of normal stem cells, such as self-renewal capacity, slow cell cycle activity, differentiation capacity, possessing an enhanced resistance towards cytotoxic agents and radiation, as well as tissue restoration capacity. Due to the increased drug and radiation resistance and slow cell cycle activity concomitant with tumor initiation capacity it is generally assumed that recurrent cancers originate from first line therapy surviving CSCs. But how does the CSC hypothesis explain "oncogenic resistance", which describes the phenomenon that most recurrent cancers are characterized by both an increased malignancy as well as resistance towards first line cancer therapy. To us, "oncogenic resistance" can not be simply attributed to the resistance properties of normal CSCs. If so, the recurring tumor should be treatable by first line therapy, which is mostly not the case. Thus, we conclude that "oncogenic resistance" demands a new type of tumor initiating cells, the so-called recurrence CSCs (rCSCs). This type of tumor initiating cell originates during first line therapy and is characterized by giving rise to first line therapy resistant and highly malignant progenies. Because several characteristics of "oncogenic resistance", such as increased drug resistance, increased resistance to apoptosis and an enhanced malignancy have been linked to cell fusion we further conclude that rCSCs might originate from this cellular event. However, which cell types have to fuse with each other to ultimately give rise to rCSCs is not clear. In any case, tumor tissues, particularly those being destructed by first line therapy comprise of a variety of fusogenic cells including tumor cells and CSCs as well as recruited monocytes/macrophages and bone marrow-derived stem cells. The fusogenic properties of these cells concomitant with phenotypic heterogeneity, which is also a property of cell fusion, will then lead to the origin of rCSCs. In accordance with Darwinian evolution only those cells will survive that can resist best to the selection pressure first line therapy.