The cancer stem cell theory, which holds that a small number of tumorigenic cells are responsible for metastases, applies only to some cancers and possibly to certain patients, according to Sean Morrison, founder of OncoMed, now at the University of Michigan Center for Stem Cell Biologics. Therefore, learning when the cancer stem cell model is valid is a vital step to understanding and, ultimately, treating cancer.
“Not every cancer cell is bad. Some cancer cells differentiate into cells that are benign,” Dr. Morrison emphasized as he opened the cancer stem cell workshop at the American Society for Cell Biology (ASCB) meeting in San Diego in December. “Cancer stem cells have the unique capability of proliferating.”
The stochastic, or clonal evolution, model dictates that, “despite the heterogeneity (in cancer), cells of many different phenotypes are capable of proliferating extensively and forming new tumors,” Dr. Morrison summarized.
The cancer stem cell model, in contrast, says that “cancer stem cells are uniquely capable of proliferating extensively and forming new tumors, and that they go through an aberrant process of differentiation giving rise to phenotypically diverse cancer cells with a limited capacity to divide.” Which of these two theories is correct seems to depend upon the type of cancer.
Researchers are trying to determine whether such tumorigenic potential is within all cancer cells, or whether it is confined to a small number of cells. “If the cancer stem cells are rare,” he said, “the ability to identify these cells is critical to studying them. If the ability to become tumorigenic is common, distinguishing them is less valuable.”
“Studies are showing that a variety of cancers follow the cancer stem cell model, where markers can be identified that distinguish cancer stem cells from nontumorigenic cancer cells,” Dr. Morrison explained. “One of the untested, but underlying assumptions of the cancer stem cell model is that the difference that distinguishes tumorigenetic and nontumorigenetic cells is epigenetic, rather than genetic in nature.”
The clonal evolution model predicts genetic changes that will be deleterious to the cell, so the observation that not every cell is tumorigenetic is consistent with the clonal evolution model. What’s different about the cancer stem cell model is the idea that you can have cancer from only a small minority of the cells, that can differentiate into nontumorigenic cells, he said.
Most of the talk about cancer stem cells is mainly predictions, he cautioned. Conclusions about the fraction of cancers that follow the stem cell model are based primarily upon markers that distinguish tumorigenetic from nontumorigenetic cells. “So, all of the conclusions depend upon the robustness of these markers,” he noted, at least some of which are less robust than initially believed. Additionally, “Do some patients follow the model and others not? Does metastasis arise exclusively from migration of cancer stem cells? That makes intuitive sense, but isn’t tested.”
“We can’t make general conclusions that cancer stems cells are less sensitive to therapy,” Dr. Morrison said. There are several potential explanations for drug resistance. For example, “the apparent resistance may reflect a log kill, in which a therapy eradicates 95 percent of the tumorigenic cells but the remaining five percent proliferate.”
In his own lab, experiments with acute myeloid leukemia and acute lymphoblastic leukemia in mice showed that not every cell proliferates and transfers disease. “Cells that express a phenotype similar to hematopoietic stem cells are 400-fold enriched for leukemigenic activity as compared to the bulk population of bone marrow cells.” That supports the idea, he said, that “not every cell has the capacity to transmit disease, and cells that transfer disease are rare.” A similar study also showed that chronic myeloid leukemia (CML) followed the cancer stem cell model.
In studies of imatinib response against CML using a different mouse model, imatinib shrunk the (enlarged) spleen to normal levels. It did not cure the disease though. In fact, cells from these mice were more capable of transferring disease, showing that the “CML stem cells are orders of magnitude more resistant to imatinib than other cancer cells in the same mice, which is consistent with clinical experience.”
Melanoma is another model that was thought to follow a cancer stem cell model. It doesn’t, Dr. Morrison reported, based upon transplanting melanoma cells from several patients into NOD/SCID mice. After eight weeks, only a minority of cells had formed tumors. At 28 weeks, however, the ratio of tumorigenic cells had increased nearly eightfold. “One issue, therefore, is that people sometimes don’t run their assays long enough to detect the full spectrum of tumorigenic cells.”