CD133+, CD166+CD44+, and CD24+CD44+ phenotypes fail to reliably identify cell populations with cancer stem cell functional features in established human colorectal cancer cell lines.

Increasing evidence that cancers originate from small populations of so-called cancer stem cells (CSCs), capable of surviving conventional chemotherapies and regenerating the original tumor, urges the development of novel CSC-targeted treatments. Screening of new anticancer compounds is conventionally conducted on established tumor cell lines, providing sufficient material for high-throughput studies. Whether tumor cell lines might comprise CSC populations resembling those of primary tumors, however, remains highly debated. We have analyzed the expression of defined phenotypic profiles, including CD133+, CD166+CD44+, and CD24+CD44+, reported as CSC-specific in human primary colorectal cancer (CRC), on a panel of 10 established CRC cell lines and evaluated their correlation with CSC properties. None of the putative CSC phenotypes consistently correlated with stem cell-like features, including spheroid formation ability, clonogenicity, aldehyde dehydrogenase-1 activity, and side population phenotype. Importantly, CRC cells expressing putative CSC markers did not exhibit increased survival when treated with chemotherapeutic drugs in vitro or display higher tumorigenicity in vivo. Thus, the expression of CD133 or the coexpression of CD166/CD44 or CD24/CD44 did not appear to reliably identify CSC populations in established CRC cell lines. Our findings question the suitability of cell lines for the screening of CSC-specific therapies and underline the urgency of developing novel platforms for anticancer drug discovery.