Nearly half of all cancer patients receive radiation treatment. It's not always effective, in part because in the process of killing off cancer cells, radiation also takes a toll on healthy cells. Researchers have recently found a way to enhance the effectiveness of radiation treatment of tumors in mice, slowing tumor growth up to nine−fold. And it works by making normal cells nearly immune to radiation's toxic effects.

The research here suggests that radiation doesn't directly kill cells but causes cells to kill themselves by sending out a self−destruct signal.

It all started in 2008, when researchers at the National Cancer Institute found that tissue death from radiation could be reduced substantially by blocking a signal that's sent within individual cells. They were studying a protein called Thrombospondin−1 (TSP1), which is produced by cells in response to inflammation and injury. When TSP1 binds to another protein at the cell's surface called CD47, a signal is sent within the cell that triggers a whole chain of events. Blocking the binding stops the signaling, preventing this chain of events from even starting. The researchers were able to produce a non−toxic inhibitor of this binding and found that one result of using it was to make cells nearly immune to the toxic effects of high radiation doses. This was seen both in mice and in cells cultured from human vascular tissue.

The high energy of radiation damages a cell's components, particularly its DNA. This has always been thought to be sufficient to explain the deadly effects of radiation on living creatures. The research here suggests that radiation doesn't directly kill cells but causes cells to kill themselves by sending out a self−destruct signal. There are many implications to this. The researchers found a practical one in the radiation treatment of cancers.

When mice are injected with melanoma (skin cancer) cells, tumors grow and the mice die within three to four weeks. If they are given an external dose of radiation one week after tumor cell injection, this slows tumor growth somewhat.

In a 2009 study, researchers found that if the mice were given a cell signaling inhibitor (here, an agent that blocked CD47 production) two days prior to radiation treatment, there was a dramatic reduction in tumor size 30 days post−radiation. Tumors in the mice given inhibitor were about one−ninth the size of tumors in the mice given radiation alone. This effect was not seen if the mice were given inhibitor but no radiation treatment.

To show that this effect wasn't something specific to this particular strain of mouse or tumor type, a similar experiment was done on another mouse strain using squamous cell carcinoma cells. Here, the tumors in the mice given inhibitor prior to radiation treatment were about one−quarter the size of the tumors in mice given radiation alone.

The researchers aren't sure why this treatment works. Use of inhibitor does not make the tumor cells more sensitive to radiation. They speculate that by leaving the mouse immune system relatively unharmed from radiation damage, the inhibitor is allowing for a better immune response to the tumor.

The radiation dose used in the tumor experiments was 10 gray (Gy), which is about five times the dose normally used in a single human cancer treatment. In other experiments, the researchers found signaling inhibitor protective against normal tissue damage at doses up to 25 Gy.

In the past, attempts to protect cancer patients from the harmful effects of radiation treatment on healthy tissue usually have also protected the cancer cells. This research suggests a way around that problem, as well as providing some insight into the nature of radiation damage itself.

An article detailing the study on delay of tumor growth in mice was published in the October 21, 2009 issue of Science Translational Medicine. The 2008 study detailing the discovery of the protective effect from radiation caused by blocking cell signaling was published in the October 2008 issue of The American Journal of Pathology and is freely available at http://ajp.amjpathol.org/cgi/content/abstract/173/4/1100.