Study Reveals Seeds of Metastases

Rare tumor cells thought to initiate metastasis possess unique gene signatures that distinguish them from cells in primary tumors, a recent study concludes. The findings may help in developing drugs that specifically target metastasis in patients with different types of cancer.

Researchers created a highly sensitive fluorescence-activated cell sorting–based assay that enabled them to isolate cells from primary breast cancers that had metastasized to different tissues after being implanted in mice. They then analyzed gene expression in these cells using a microfluidics-based platform. They found that metastatic cells from tissues with a high metastatic burden were similar to those in the primary tumor, whereas cells from tissues with lower metastatic burden (early stage) had a distinct gene signature.

“We discovered that particularly early metastatic cells have a distinct phenotype that is more stem cell-like and distinct from mature metastatic cells,” says senior author Zena Werb, PhD, professor and vice-chair of anatomy at the University of California, San Francisco. “These early metastatic cells appear to initiate metastases and trigger advanced metastatic disease.”

The researchers developed a method that uses flow cytometry and a humanized antibody to extract individual human metastatic cells from xenograft mice, says lead author Devon Lawson, PhD, assistant professor of physiology and biophysics at the University of California, Irvine.

“We discovered how to fish out cells as they are metastasizing and view them at very high resolution using single-cell technology,” says Lawson. “We could see 10 individual human cells in the lung, for example, because the antibody is so specific to human cells.”

A subset of primary tumor cells possess a stem-like phenotype, which is consistent with the team's previous findings that “leader” cells on the periphery of primary breast tumors express basal-cell markers associated with initiation of metastasis. The researchers also discovered that the more stem-like cells a tumor contained, the greater its potential to metastasize.

“We don't know whether these cells are propagating independently of more differentiated ones,” says Werb. “However, by looking at human cancer databases, we learned that the tumors that metastasize express—at much higher levels—the same subset of genes that we found being expressed on the periphery of breast tumors.”

In addition, early-stage metastatic cells tended to express higher levels of dormancy-associated genes, whereas later-stage cells expressed higher levels of cell cycle–promoting genes, such as MYC and CDK2, which have been associated with reactivation after dormancy.

Based on the findings that early-stage metastatic cells appear to upregulate MYC, the researchers tested whether dinaciclib, a CDK inhibitor, would induce cell death. After 4 weeks of treatment, only 1 of 24 treated animals displayed metastatic cells, compared with 11 of 25 controls. Growth of the primary tumor was delayed but later progressed in drug-treated animals, suggesting that the drug specifically targets metastatic cells.

Studying gene expression in individual metastatic cells may help identify new potential drug targets, says Werb.

“This study provides a tractable assay for looking at targets for metastases, which cause the majority of cancer deaths,” she says. “If we can identify genes involved in these early events, we may eventually be able to identify patients with the worst prognoses and avoid overtreatment of other patients.”