As outlined herein, there is now much evidence that increased Rel/NF-κB signaling contributes to human cancer, and that this pathway will continue to receive attention as a promising molecular target for cancer therapy and prevention. However, there remain many molecular details to resolve. Moreover, at least in some cases, constitutive p50-RELA activity may be an adaptation of certain tumor cell lines to growth in tissue culture or may be a symptom of an abnormal, tumor-induced differentiation program. For example, Cogswell et al. (2000) found that primary human breast cancer tumor cells have active p52, REL, and BCL-3, whereas breast cancer cell lines have constitutively active RELA. Consistent with that finding, over-expression of RELA reduces the tumorigenicity of one breast cancer cell line in vivo (Ricca et al., 2001). Finally, given that the Rel/NF-κB pathway can have opposite effects on growth and apoptosis in different cell types, all cancers may not respond in the same way to inhibition of the NF-κB pathway. Indeed, overexpression of the IκBα super-repressor promotes skin carcinomas in one transgenic mouse model system (van Hogerlinden et al., 1999). Thus, the growth and survival of a given tumor cell type is likely to depend on a balance between the activity of the Rel/NF-κB pathway and the activity of many other signaling pathways in ways that are not always easy to predict. In addition, as documented in this collection of articles, it is likely that the Rel/NF-κB pathway is only one of several signaling pathways that are commonly activated in human cancers