Targeting Cancer Stem Cells

By Allison Proffitt

August 13, 2009 | Researchers have been able to use a special screening technique to identify an agent that specifically kills cancer stem cells. The team, led by Piyush Gupta, Broad Institute; Tamer Onder, Whitehead Institute; Robert Weinberg, Whitehead Institute; and Eric Lander, Broad Institute; will publish their findings in the Aug 21 issue of Cell.

Cancer stem cells (CSCs) are a subset of tumor cells responsible for driving tumor growth and recurrence. The stem cells "are resistant to many existing anti-cancer therapies," explains Gupta. Even when traditional treatment kills the bulk of tumor cells, CSCs can "survive to regenerate new tumors,” said the study authors. “CSCs have [also] been proposed to be responsible for colonization at secondary organ sites upon metastatic dissemination,” the paper continues.  

Screening specifically for agents lethal to CSCs, though, has been difficult. “Failure to find such agents before has largely been because screens have looked at bulk populations, but CSCs are a small minority,” Eric Lander tells Bio-IT World. In this case, “molecular biological methods were used to ‘lock’ cells into a stem-like state, allowing screening,” he says. Researchers induced epithelial-mesenchymal transition (EMT) in breast cancer cells, which increased the number of CSCs by 100 fold, and made screening possible.

Screening 16,000 compounds, the researchers found 32 that exhibited at least some selective toxicity to the CSCs. Eight of these were tested further, and one compound, salinomycin, was found to be selectively toxic to breast cancer CSCs with 8-fold selectivity, much higher than the other compounds. In one experiment, salinomycin showed a 75-fold reduction in CSCs compared to the control. In another, the agent seemed to decrease tumor-seeding ability by 100-fold.

Lander stressed that salinomycin itself is not as interesting as the process. “The particular agent is not likely to be a drug itself, but the proof of principle is important,” he said. “The paper shows that it is actually possibly to find drugs that target [CSCs].”

The study authors concede that “non-CSCs within tumors can give rise to CSCs at low but significant rates,” and that “the elimination of the CSCs within the tumor may not result in its complete regression,” but the ability to target CSCs selectively represents a critical step forward.

The authors suggest that in the future the screening technique may help researchers identify agents that target both CSCs and non-CSCs, or formulate combination therapies with toxicity toward each cell type.

Robert Weinberg believes that this is an important first step toward targeting this tumor cell type. “Since CSCs represent the source of regrowth of tumors once they've been reduced in size, and since CSCs are more resistant to killing by a number of conventional chemotherapeutics, it seems likely that over the next decade many will come to realize that it will be necessary to target the CSCs in addition to ‘debulking’ the tumor with conventional chemotherapeutics that target the non-CSCs within a tumor,” he told Bio-IT World.