Biomedical Engineering Reference
In-Depth Information
expression profile associated with CSCs (Balic et al. 2006). This strongly sug-
gests existence of mCSCs capable of proliferating in the bone marrow, the most
common site of breast cancer metastasis. It should be noted, however, that
pioneering studies to identify pCSC markers in breast cancer assayed cells from
pleural effusions, not primary tumors (Al-Hajj et al. 2003). It is interesting that
the CD44
+
/CD24
surface marker signature seems to correlate both with
pCSC identification as well as with metastatic characteristics. CD44
+
/CD24
cells exhibit highly invasive phenotypes in vitro (Sheridan et al. 2006) and
overexpress several genes involved in chemotaxis and motility (Shipitsin et al.
2007). The invasive gene signature found in these cells may be a result of
inherent characteristics of the population studied (owing to isolation from
outside the primary tumor) or may suggest that pCSC and mCSC populations
in some breast cancers are overlapping.
As evident in breast cancer, much work is needed to identify characteristics
and origins of mCSCs. Direct confirmation of mCSCs as a unique cell type has
only been accomplished very recently in a single pancreatic cancer model (Her-
mann et al. 2007). A logical approach to this problemwould be to use the markers
that have identified pCSCs and test them in metastatic assays. Xenograft trans-
plantation of fractionated metastatic tumors or circulating tumor cells could
identify mCSC cells by their capability to serially induce tumor formation. This
fraction could then be assayed for metastasis-related phenotypes such as ability
to colonize various tissues like bone, lung, brain, and other clinically relevant
sites. Once a tumorigenic metastatic group is isolated, it can be fractionated based
on surface markers and gene expression profiles, yielding a wealth of data critical
to the understanding of metastasis. While recent research has focused on end-
points of metastasis measured by expansion of the secondary tumor, other steps
of the metastatic cascade must affect mCSCs as well. Processes of angiogenesis,
invasion into surrounding stroma, intravasation into lymphatics or blood vessels,
adhesion to luminal epithelia, and extravasation should all be explored with
regards to mCSC function. This would provide insight into mCSC function
and tissue tropisms as well as the process of metastasis in general. Understanding
tumor dissemination at a mechanistic level, especially with regards tomCSCs will
reveal new potential intervention points for treatment as well as new criteria for
diagnosis. Some progress has been made toward this end in angiogenesis induced
by mCSCs in glioma. CD133
+
CSCs are more capable in inducing angiogenesis
than their CD133
counterparts (Bao et al. 2006b) suggesting CSC involvement
in an early step of metastatic progression and potential for involvement in
subsequent steps of the cascade.
5 The Pre-metastatic Microenvironment and mCSCs
Non-cell-autonomous interaction with the environment has long been known
to play a key role in cancer metastasis. Signaling interactions with surrounding
cells and manipulations of adhesion allowing modified association with the
