New cellular target may put the brakes on cancer’s ability to spread

May 30, 2017
Histopathology of urothelial carcinoma of the urinary bladder

Histopathology of urothelial carcinoma of the urinary bladder (Image: KGH)

A team led by Johns Hopkins researchers has discovered a biochemical signaling process that causes densely packed cancer cells to break away from a tumor and spread the disease elsewhere in the body.

In their study, published online today in Nature Communications, the team also reported that the combined use of two existing drugs disrupts this process and appears to significantly slow cancer’s tendency to travel, a behavior called metastasis.

The new findings are important, the researchers said, because 90 percent of cancer deaths are caused by metastasis, and anything that derails this activity could improve the prognosis for patients. The crucial new signaling process turned up when the team took a closer look at cellular events that promote metastasis.

“We found that it was not the overall size of a primary tumor that caused cancer cells to spread, but how tightly those cells are jammed together when they break away from the tumor,” said lead author Hasini Jayatilaka, a postdoctoral fellow at Johns Hopkins’ Physical Sciences-Oncology Center. “At a fundamental level, we found that cell density is very important in triggering metastasis. It’s like waiting for a table in a severely overcrowded restaurant and then getting a message that says you need to take your appetite elsewhere.”

Jayatilaka and her colleagues found a medication mix that stopped delivery of this microscopic message. The team members cautioned that this treatment was tested in animal models, but not yet on human cancer patients. Nevertheless, they said the discovery contributes to a promising new focus for cancer research: disrupting the biochemical activity that prods cancer cells to spread through the body.

One of the study’s senior authors, Denis Wirtz, Johns Hopkins University’s vice provost for research and director of its Physical Sciences-Oncology Center, said no commercial drugs are now being produced specifically to inhibit metastasis because drug companies believe the best way to stop the spread of cancer is to destroy the primary tumor from which it originates.

“The pharmaceutical companies view metastasis as a by-product of tumor growth,” said Wirtz, who also holds Johns Hopkins faculty appointments in chemical and biomolecular engineering, in pathology, and at the Johns Hopkins Kimmel Cancer Center. “Our study looked more closely at the steps that actually initiate metastasis. By doing this, we were able to develop a unique therapeutic that directly targets metastasis, not the growth of the primary tumor. This treatment has the potential to inhibit metastasis and thus improve cancer patient outcomes.”

Excerpted from The Hub.

Back to top