The cancer stem cell (CSC) model is one that is utilized to explain the cellular heterogeneity that exists within tumors. Max Wicha, MD, the Madeline and Sidney Forbes Professor of Oncology and Founding Director Emeritus, University of Michigan Comprehensive Cancer Center, gave a presentation highlighting some recent advances within that field.
Wicha noted that “many survivors of the Hiroshima and Nagasaki bombings did not develop breast cancers until decades after exposure to the blasts' atomic radiation. That suggested that the cancerous cells were decades old, thus implying an association of the cancer with adult stem cells, as those cells are the only ones with such long lifetimes.” Additionally, he said “The highest cancer rates were in pre- and peripubescent girls, a population which would be expected to have elevated adult stem cell activity.”
Studying Cancer Stem Cells
In the CSC model, there exists a hierarchical cellular order within a tumor, with the top of this order being occupied by cells that exhibit stem-cell like properties, i.e., they can self-regenerate or produce differentiated daughter cells that then make up the bulk of the tumor. Such hierarchies have been noted in cases of human leukemia as well as in solid tumors of the following cancers: breast, brain, prostate, pancreas, colon, lung, and ovarian. Evidence for the existence of these CSCs in the case of breast cancer was presented in a seminal 2003 publication by Al Hajj, et al, (doi: 10.1073/pnas.1131491100). In their work, the authors isolated and identified the tumorigenic cells as those expressing CD44+ CD24-/low cell surface markers. With as few as hundreds of these cells, they were able to produce tumors in immunocompromised mice. When tumor cells of different phenotypes were utilized, injection of up to 10,000 cells failed to initiate tumor growth in similar mice. When the CD44+ CD24-/low cell initiated tumors in the mice were analyzed, the causative CD44+ CD24-/low cells were observed, however, the bulk of the tumors consisted of cells having non-tumorigenic phenotypes.
This observation clearly showed that the tumorigenic cells had differentiated into those having a non-tumor initiating phenotype. Wicha noted, “Additional cellular diversity can be obtained via a process of ‘dedifferentiation’, in which, differentiated cells can acquire a more stem-cell-like character. These dedifferentiated cells can then undergo subsequent differentiation to cells of differing phenotypes.” All cells, however, do not have the same ability to undergo this process; the more differentiated a cancer cell is, the less likely it is to undergo a subsequent dedifferentiation.
In addition to the CD44+ CD24- cells, breast cancer cells expressing the enzyme aldehyde dehydrogenase (ALDH1) as well as those being CD44+ CD24- /ALDH+ have also been shown to be tumorigenic (DOI: 10.1016/j.stem.2007.08.014). The CD44+ CD24- cells are termed mesenchymal-like, (undergoing an epithelial-mesenchymal transition (EMT)) and tend to be quiescent but invasive, and are located at the interface of the tumor, while those cells expressing ALDH are termed epithelial-like (undergoing mesenchymal-epithelial transition (MET)) and are more proliferative and are found at the center of the tumor. Those cells displaying CD44+ CD24- /ALDH+ were shown to be the most tumorigenic, however, it is not known if this is a transient or stable cell type.
“With these differing cell types, one can easily see a mechanism for metastasis. Cells at the tumor interface transition to an EMT (mesenchymal) state and are transported to distant sites, where they are implanted and undergo a subsequent transition to an MET (epithelial) state and become proliferative,” Wicha observed.
An interesting result was obtained by researchers investigating the case of increased trastuzumab (anti-human epidermal growth factor type 2 (HER2) monoclonal antibody) resistance in HER2+ breast cancer cells (DOI: 10.1016/j.molcel.2012.06.014). There, the researchers showed that by having “knocked down” PTEN expression in HER2 overexpressed cell cancer lines, they were able to induce trastuzumab resistance in the cells via an interleukin-6 (IL-6) feedback loop.
As Wicha explained, “The IL-6 feedback loop caused an expansion in the number of EMT cancer stem cells. Additionally, these cells secreted more than 100 times the IL-6 of the parent cells.” The use of an IL-6 receptor antibody in mouse xenograft models disrupted this feedback loop, decreasing the numbers of cancer stem cells present, thus resulting in reduced metastases and tumor growth. This finding suggests a possible clinical strategy to circumvent trastuzumab resistance.
This observation shows that cytotoxic therapies can often increase the number of Cs in a patient via activated cytokine feedback loops (e.g., IL-6 and IL-8). One important fact to note is that in many instances, the patient's immune system does have a profound effect on the tumor microenvironment. Many mediators of EMT and MET are immune-system associated in nature, and thus are absent in the immunocompromised mice utilized in xenograft studies. “The ideal model for mimicking the tumor microenvironment would be an immunocompetent humanized mouse model,” Wicha commented.
With regards to future directions for this promising research, Wicha noted, “Currently, we assess a drug's efficacy by its ability to reduce the bulk tumor's size; however, this clearly does not always correlate with positive patient outcomes.” He continues “The most effective therapy would most likely utilize an approach which targets the EMT and MET phenotype CSC's as well as the non-tumorigenic bulk cancer cells.”
Wicha received the 9th annual AACR Distinguished Lectureship in Breast Cancer at the 2016 San Antonio Breast Cancer Symposium.
Richard Simoneaux is a contributing writer.