The largest challenges facing the usage of stem cell are sourcing and safety.  Although the complete pluripotency of embryonic stem cells has been established, the moral issues surrounding their usage and their general unavailability to current patients likely precludes their widespread clinical usage.  The majority of groups currently studying the usage of transduced stem cells as vehicles for transgene products utilize adult neural stem cells.  However, the derivation of neural stem cells from living patients is currently not possible.  As a result, more efforts needs directed toward the utilization of adult mesenchymal stem cells.  As mentioned in the stem cell migration section, several groups have used bone marrow-derived stem cells effectively and this avenue of research demands more work.

Alternatively, it is possible that sourcing issues may be ameliorated by usage of induced pluripotent stem cells.  Although this sourcing method averts moral objection and theoretically could provide large numbers of autologous stem cells, the methods used for de-dedifferentiation are still not well understood.  Although it has been found that this method can be used without the usage of oncogenes, the formation of teratomas in vivo makes adoption of this technique unlikely before greater understanding of the mechanisms controlling differentiation and de-differentiation is achieved. 

It has been suggested that the creation of de novo tumors occurs via one of three mechanisms: expression of the gene used for immortalization of delivery cell line, insertion of the immortalizing or therapeutic transgene into a critical genomic location causing deregulation of protein expression, or the fusion of immortalized delivery cells with endogenous cells resulting in a new cell with a transformed phenotype.  It is clear that a possible way to avoid these complications is to use stem cells which have not been immortalized, however, this confounds cell sourcing further and also limits the survival time of stem cell upon implantation.        

Of additional concern is immune response due to allogeneic stem cell transplantation.  Although it has been shown that central nervous system stem cells do not express major histocompatibility complex class I or II antigens in their undifferentiated state and thus invoke a limited immune response, implantation of large numbers of cells had yet to be characterized in terms of immune response.(33)  Should immune response be significant, however, treatment may still be viable as it has been shown that immune suppression in mice using cyclosporin during treatment with allogeneic stem cells did not affect therapeutic effectiveness or tumor-tropic properties of stem cells.