Even when anti-cancer drugs appear to eliminate tumors, a resistant few malignant cells can survive and allow cancers to resume a fatal onslaught. Researchers at OHSU in Portland have revealed a previously unknown process used by breast cancer cells to shift rapidly into a drug-tolerant state – and they’ve demonstrated a successful strategy to combat it.
“It’s a different way of thinking about drug resistance,” says Rosalie Sears, Ph.D., senior author of the paper describing the work. Sears is a professor of molecular and medical genetics in the OHSU School of Medicine and co-director of the Brenden-Colson Center for Pancreatic Care.
Image: a breast cancer cell viewed by scanning electron microscopy (NCI Visuals Online)Resistance to anti-cancer drugs emerges by more than one mechanism. Mutations, for example, can change the shape of a cancer protein so that a targeted therapy drug can no longer interact with it. Cancer cells can evolve to pump out drugs or inactivate them, to crank up repairs of DNA damaged by drugs, or to use new growth-signaling pathways to work around those blocked by targeted therapy drugs.
For the most part, drug resistance was thought to emerge via Darwinian selection, in which random mutations equip a few cancer cells with a survival advantage, allowing these cells to multiply and spread despite anti-cancer treatment.
The new study shows how new mutations aren’t needed for some breast cancer cells to rapidly change into a drug-tolerant state. The cancer cells reprogram particular sets of genes so that some become active and others quiescent in a way that makes them impervious to the effects of an anti-cancer drug. The changes occur in a matter of hours. The findings add to mounting evidence pointing to epigenetic remodeling as a critical driver of drug resistance.
“Cancer cells can go into a disguised state and wait it out until the pressure is gone,” says Ellen Langer, Ph.D., second author of the paper. “What we’ve shown in no way discounts the fact that Darwinian selection occurs. It’s more the idea that even on top of this clonal selection there is another mechanism of resistance that doesn’t require additional mutations.” Langer is a research assistant professor in the Department of Molecular and Medical Genetics. The paper was published in Nature Communications.
‘If we prevented cancer cells’ ability to switch into a new differentiation state, targeted therapy was able to get rid of the whole population of tumor cells.’
The researchers began their investigation by using a set of three proteins, vimentin and two types of keratin, to sort out the states of tumor cells obtained from patients with breast cancer. The presence of vimentin and absence of keratin indicates cells that have transitioned to a so-called mesenchymal state. Absence of vimentin but presence of keratins indicates cells in a basal or luminal state, depending on the type of keratin. They found that in aggressive forms of breast cancer, tumors contained diverse mixtures of cells in every differentiation state.
Using laboratory-grown breast cancer cell lines, the researchers studied how tumor cells respond to a variety of targeted-therapy drugs. (The cell lines contained cells in all of the differentiation states: luminal, basal, and mesenchymal.) Targeted therapy drugs generally killed many tumor cells but left some survivors. The surviving cells were reprogrammed into a drug-tolerant persister state defined by minimal cell division and an altered pattern of gene expression. This state could persist for weeks in the face of high doses of anti-cancer drug.
Contrary to what would be expected with Darwinian selection, death and replacement of cancer cells was not required for a drug-tolerant persister state to emerge following treatment. Gene sequencing of surviving cells showed no evidence of Darwinian selection. And after drug withdrawal, cancer cell populations regained a diverse mix of differentiation states and drug sensitivity comparable to an untreated population of cancer cells.
The findings suggest that when exposed to a targeted therapy that inhibits one vital signaling pathway, breast cancer cells can survive by transitioning to a differentiation state that doesn’t need that signaling pathway. “We started with one therapy, they went to one state. Another therapy, they went to another state. And then we combined those therapies thinking that might work, but they went to even a third state. So they seem to be able to just switch signaling pathway usage to survive,” Sears said.
Further experiments showed that the cancer cells that survived targeted therapy showed high levels of chromatin remodeling, which is how cells open up sections of chromosomes for the switching on of particular genes. And chromatin remodeling is necessary for cells to transition between differentiation states.
That suggested a strategy to combat resistance by adding a second drug to block chromatin remodeling. The researchers used a drug called a BET inhibitor, which blocks proteins involved in chromatin remodeling. The results were dramatic.
“Without the BET inhibitor, any combinations of targeted inhibitors we used always left drug-tolerant persisters that could come back and were resistant to anything we threw at them,” Sears says. “But if we prevented cancer cells’ ability to switch into a new differentiation state with the BET inhibitor, targeted therapy was able to get rid of the whole population of tumor cells.”
Those experiments were in cancer cell lines. In animal models, the combination of BET inhibitor and targeted therapy drugs wasn’t as effective. “The tumors regressed but they didn’t disappear so it’s not the end of the story,” Langer says. “Their continued resistance is probably linked to something in the microenvironment that’s allowing them to get to a state where they can still survive.”
To advance the research, Langer and Sears say they will test different BET inhibitors and related drugs to find ones that work the most effectively in animal models to stop cancer cells from transitioning into drug-tolerant states. The researchers are also planning experiments to gain a deeper understanding of the precise steps that cancer cells follow in transitioning to new differentiation states, and the microenvironment factors that support their ability to transition.
“If we can understand the epigenetic remodeling that underlies the ability of cells to go to these really resistant cell states, then we can hopefully learn how best to treat them,” Sears says.
Differentiation-state plasticity is a targetable resistance mechanism in basal-like breast cancer by Tyler Risom, Ellen M. Langer, Margaret P. Chapman, Juha Rantala, Andrew J. Fields, Christopher Boniface, Mariano J. Alvarez, Nicholas D. Kendsersky, Carl R. Pelz, Katherine Johnson-Camacho, Lacey E. Dobrolecki, Koei Chin, Anil J. Aswani, Nicholas J. Wang, Andrea Califano, Michael T. Lewis, Claire J. Tomlin, Paul T. Spellman, Andrew Adey, Joe W. Gray and Rosalie C. Sears. Nature Communications (September 2018)