Growing knowledge concerning tumor heterogeneity has increased the importance of analyzing tumor resident cell and stem cell subpopulations. In particular, cancer stem cells (CSCs), also called tumor initiating cells, have gained substantial interest in the research field over the past few years. CSCs have been isolated from multiple tumor entities and were shown to play a crucial role during tumor growth and metastasis. However, isolation and analysis of CSCs even from single tumor entities like melanoma showed contradictory results among different research groups. This can at least partially be explained by variations in the dissociation and isolation methods used. One critical point that has to be considered during the analysis of putative CSC markers is that the procedure used for dissociation of primary tissue prior to staining may cause a technical bias. Aggressive proteases, such as trypsin, which cleave off several CSC surface markers like ABCB5 or CD44, may cause incorrect results concerning the differential expression of sensitive markers (Civenni et al., 2011). A further source of non-reproducible findings is the use of antibodies that are specific to sub-forms of cell surface markers. Glycosylation and splice-isoform dependent epitopes, as found for e.g. CD24 and CD133, are differentially expressed among tissues and cell states. In particular, it was shown that the AC133 epitope but not the entire CD133 protein expression is lost upon CSC differentiation (Kemper et al., 2010). We have developed a standardized platform that allows fast and highly reproducible CSC enumeration and isolation from various tumor tissues. This platform includes automated and therefore user independent procedures for dissociation of human and implanted mouse tumor tissue without degradation of cell surface epitopes yielding high amounts of single cells at viabilities around 90%. Furthermore, antibody conjugates of clones recognizing epitopes shown to be relevant for CSC analysis were generated and tested on cell lines as well as primary tumor tissue allowing for the quantification and isolation of CSCs. As a proof of principle, highly viable populations of CD133+ and CD45/CD24/CD44+ CSCs were efficiently isolated from a glioblastoma at 86% and an invasive ductal mamma carcinoma at 94% purity, respectively. In addition, optimized gating and exclusion strategies are presented avoiding analytical bias for flow cytometry-based CSC quantification from primary tissue as shown for human melanoma tissue. By using reliable methods for the analysis of CSCs from various origins, a major cause of conflicting results in this field may be minimized and comparisons of these subpopulations among distinct tumor entities will be possible.

Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 103rd Annual Meeting of the American Association for Cancer Research; 2012 Mar 31-Apr 4; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2012;72(8 Suppl):Abstract nr 5340. doi:1538-7445.AM2012-5340