Researchers Identify How Nongenetic Mechanisms of Cancer Cell Diversity Can Affect Treatment Response

A new study suggests that even tumor cells with a common genetic background can display functional heterogeneity. The results point to nongenetic mechanisms, such as microenvironment and epigenetic differences, that contribute to tumor growth as well as tolerance to therapy. The research, led by Antonija Kreso, PhD, Catherine O’Brien, MD, and John Dick, PhD, of the Princess Margaret Cancer Centre and the University of Toronto, Canada, was published in the February 2013 issue ofScience.

“Genetic mutations are thought to account for tumor formation and growth, but our results indicate intra-clonal diversity with respect to functional properties also plays a role,” said Kreso.

The predominant theory of solid tumor heterogeneity states that tumor cell diversity is a result of progressive mutational changes that form distinct tumor cell clones, with each type of subclone thought to be genetically similar. But how nongenetic factors contribute to tumor heterogeneity along with genetic diversity is not clear.

Kreso and colleagues addressed whether these genetic clones are functionally equivalent and respond in the same manner to chemotherapy treatment by tracking individually labeled human colorectal cancer (CRC) cells in mice. They found that individual tumor cells within the same lineage had a spectrum of survival and growth patterns and responded differently to chemotherapy. The findings suggest that there are more mechanisms that contribute to cancer treatment failure than simply acquired genetic mutations of the tumor.

To understand whether the cells of a tumor that drive tumor growth are genetically the same, and whether these cells could have different functions, the authors both dissected the genetic sequences of the cells and performed functional tests on individually marked cells to measure how well each cell could drive tumor growth. Fluorescently labeled CRC cells derived from 10 human colorectal tumors were transplanted into mice and allowed to grow into a new tumor with the same genomic makeup as the original, tumor- initiating cells. Cells from these tumors were then serially transplanted, up to five times, into other mice and allowed to form new tumors. Each time, the genome of the CRC tumor remained stable. The time to form a stable tumor was the same over serial transplants, and each clone retained characteristics of the original human tumor.

The authors found that genetics could not explain the different behaviors of the tumors. The genetic clones that were selected in the mouse models remained very stable, yet the tumor cells displayed markedly different types of growth behaviors. While some cells divided rapidly when transplanted into new mice, others did not reproduce well through serial transplantations. Still other cells did not grow well upon a first or second serial passage, but then regained the ability to grow well upon subsequent passages. A total of five distinct functional behaviors were observed among the clones. “The distinct proliferative kinetics of the five clonal behaviors we observed underscore the functional variability of individual cells,” the authors wrote. Both deep sequencing of mutational hotspots and copy number alteration analysis were used to assess genomic diversity.

When the transplanted tumor cells were exposed to chemotherapy, they also exhibited a range of distinct responses. Mice with established CRC tumors that were marked to distinguish single cells were treated with the chemotherapy oxaliplatin, and these chemotherapy-exposed cells were then transplanted into a new mouse. The resulting tumors were generated predominantly by previously slow-growing and dormant clones, similar to the behavior of the nonchemotherapy-exposed cells in the original experiment. This suggests that chemotherapy may select for the survival of those cells that are slow-growing or even dormant. These cells can then re-initiate tumor growth posttreatment, which could account for disease recurrence after initial response to chemotherapy.

“The biggest surprise was when we found that quiescent or slowly cycling cells survive chemotherapy better than their highly proliferative counterparts,” said Kreso. These chemotherapy-exposed cells still had very similar genetic profiles to the original, nontreated cells, suggesting a nongenetic way in which cells can outlive anticancer therapy.

The genetic variability of subclones within a tumor contributes to tumor growth and survival. Different types of stress and exposure to anticancer therapy can select for the growth of subclones that are able to overcome treatment, resulting in outgrowth of tumor cells and contributing to disease recurrence. But while researchers have presumed that genetic driver mutations—those either selected or acquired—result in therapy resistance, acquired driver mutations may not always be associated with tolerance or resistance to a therapy, according to the new study.

Clinical Pearls

• Researchers have been able to track the growth of single human tumor cells into tumors over time in mice
• Tracking human colorectal cancer cells, the researchers were able to show that not just genes, but other biological factors can contribute to the growth and behavior of the colorectal tumor cells and contribute to failure of therapy and cancer relapse
• The results challenge the traditional notion that the properties of tumor cells, including resistance to treatment, are dictated only by the genetic makeup and spectrum of tumor mutations

Kreso and colleagues are now working to understand the nongenetic factors that play an important role in tumor cell behavior, especially response to chemotherapy. “We want to identify the epigenetic or environment factors that influence [a cell’s functional properties],” Kreso explained. One approach is to compare the epigenetic marks between cells with different growth kinetics observed in this study. “If we can gain a better understanding of the factors that control cancer cell behavior, we may be able to design more specific therapies.”

The researchers are also determining both the number and the functions of genetically different clones, as well as the heterogeneity between the different clones. Kreso believes that the basic principles observed in their analysis are likely to extend to other epithelial cancers.

These results have implications about how to identify novel anticancer targets. “We have to broaden our search and look for more than just genetic mutations to target tumor growth,” said Kreso. Techniques to test novel therapies also need to move beyond in vitro cell line growth assays to testing the ability of tumor cells to form new tumors. Particularly, identification of ways to prevent metastasis remains a major hurdle in cancer treatment. “It will be interesting to see whether some tumor cells are more efficient at forming metastasis, whether this is linked to the cell types we have described, and which genetic or epigenetic factors contribute to this behavior,” said Kreso.

Kreso A, O’Brien CA, van Galen P, et al. Variable clonal repopulation dynamics influence chemotherapy response in colorectal cancer.Science