1. ¹Clinic for Stem Cell Transplantation, University Medical Centre, Hamburg, Germany
2. ²Experimental Paediatric Haematology and Oncology, Johann Wolfgang Goethe University, Frankfurt, Germany
3. ³Institute for Medical Informatics, Statistics and Epidemiology, University of Leipzig, Leipzig, Germany

Correspondence: Professor Dr B Fehse, Experimental Paediatric Haematology and Oncology, Paediatric Clinic III, University Hospital, Johann Wolfgang Goethe University, Theodor-Stern-Kai 7, Frankfurt 60590, Germany. E-mail: b.fehse@kinderkrebsstiftung-frankfurt.de

Received 11 July 2007; Revised 30 August 2007; Accepted 14 September 2007; Published online 1 November 2007.

Abstract

Improvements of (retroviral) gene transfer vectors, stem cell isolation and culture techniques as well as transduction protocols eventually resulted not only in the successful genetic modification of cells capable of reconstituting the haematopoietic system in various animal models, but also human beings. This was a conditio sine qua non for the successful application of gene therapy for inherited diseases as meanwhile achieved for severe combined immune deficiencies (SCID-X1, ADA-SCID) and chronic granulomatous disease (CGD). Unexpectedly, in long-term animal experiments as well as in the follow up of patients from the CGD trial, haematopoietic clones bearing insertions in certain gene loci became dominant, which was most apparent in the myeloid blood compartment. Accumulating data strongly suggest that this clonal dominance was due to some growth and/or survival advantage conferred by gene-activating or -suppressing effects of the integrated retroviral vector (insertional mutagenesis). Importantly, such induced clonal dominance seems not to lead to malignant transformation of affected cell clones inadvertently. The latter finding has become the basis for the concept of 'induced haematopoietic stem cells', a potentially powerful tool to investigate genes involved in the regulation of mechanisms underlying competitive advantages of stem cells, but also in the multi-step nature of malignant transformation. Here we discuss promises and open issues of this concept as well as the important question of common insertion sites statistics and its pitfalls.