In this era of molecular medicine, controversy among cancer researchers is increasing as to whether the laboratory cells they study -- and upon which human treatment is based -- accurately reflect the biology of “real” tumors growing in a person’s body. Some argue that cancer cells that learn to live in a flat lab dish cannot reflect cancer in the body, but others say that without any other way to study cancer, they seem to have performed well.
Now, researchers at the Lombardi Comprehensive Cancer Center report in the December 2006 (available online November 1) issue of the International Journal of Oncology that the molecular profiles seen in a group of heavily used breast cancer laboratory cell lines significantly resemble those found in human tumors.
“We have provided an answer to this dispute, at least for cell lines that represent a majority of breast cancer cases,” said the study’s lead author, Robert Clarke, Ph.D., D.Sc., a Professor of Oncology and Physiology & Biophysics at Georgetown University Medical Center.
“Researchers -- and by extension, breast cancer patients -- can now have more confidence in these laboratory cell line models, which they use as a basis to understand the disease and design new therapies,” Clarke said.
The research team, which includes scientists from Scotland and Virginia, specifically found that three popular laboratory cultures of estrogen-sensitive breast cancer (which represents about 70 percent of the disease) share a very similar genetic profile to tumors extracted from human breasts.
The finding is important because breast cancer researchers are now using the long-existing laboratory cell lines to tease out the specific genes and proteins that are important to both development and treatment of the disease.
These lines (MCF-7, T47D, ZR-75-1) were created decades ago -- one is more than 30 years old -- from cells collected from the lungs of several unidentified women whose breast cancer had metastasized, Clarke said.
“The breast cancer had started growing in the lungs, and cells from the tumors shed into lung fluid, which was then collected,” Clarke said. These cell lines are “immortal” -- scientists can keep them growing for as long as needed, and the original population has been subdivided countless times.
But researchers have worried that this method of collection carried with it some flawed assumptions, such as the notion that because the cells had come from a tumor that had metastasized, they were also equally capable of spreading.
“We now know that is not accurate,” he said. “Cancer cells may metastasize as clumps, but not all the cells in these clumps are the same.” Separating cancer cells that spread from those that don’t is important in designing the most effective therapies, and in understanding the basic biology of the disease, according to Clarke.
And even if the breast cancer cells collected from lung fluid were capable of metastasizing, “in laboratory culture, they can lose those properties, because there is no selection pressure to retain the ability to spread,” he said. “Cells are stimulated by their environment, and those that grow on plastic won’t fully reflect what is growing in the breast.
“These cell line models can be misused if you expect them to offer biological insights into how breast cancer behaves. That is where it gets controversial,” Clarke said.
In the study, which used a new method to gauge molecular similarities between tumor cells, the scientists compared the three estrogen-receptor positive (ER+) laboratory cell lines with more than a dozen tumor biopsies that were flash frozen just after they were taken from a breast cancer patient.
They compared the cells’ “transcriptome,” the set of messenger RNA molecules being produced or active when the tissue was frozen. “This shows exactly which constellations of genes were in the process of making proteins,” he said. “This is the first time someone has looked at the question in this way, and we found the transcriptomes were not identical, but that they were surprisingly alike.”
They identified a group of 36 genes with an activation profile that was similar between the cell lines and the biopsy samples, and the researchers say that a number of these gene functions have been associated with treatment outcomes.
“The strong correlation we see between the respective transcriptomes clearly imply these laboratory cell lines are good models in which to identify molecular events that are important in some ER positive breast cancers,” Clarke said.
The study was funded in part by the United States Army Medical Research and Materiel Command Breast Cancer Research Program. Clarke’s co-authors include, from the Lombardi Comprehensive Cancer Center: Yuelin Zhu, M.D., Antai Wang, Ph.D., Minetta C. Liu, M.D., Alan Zwart, M.S., Richard Lee, Ph.D., and Ann Gallagher, R.N. Researchers from Virginia Polytechnic Institute and State University (Yue Wang, Ph.D.) and from the University of Edinburgh in Scotland (William R. Miller, Ph.D, J. Michael Dixon, M.D.) also contributed.
Cite This Page: