Bone marrow continually makes blood stem cells, which turn into new blood cells to replace spent ones, but the process is not perfect: Some blood stem cells can develop into abnormal versions, although the immune system usually stamps them out. In acute myeloid leukemia, however, the immune system seems unable to recognize malformed blood cells, which proliferate quickly and become cancerous.

Researchers from Stanford University recently uncovered a mechanism by which these leukemia cells elude the immune system. As described in two papers in Cell, the cancerous pretenders use the same "ID" that blood stem cells use to travel around the body. Their findings may not only have implications for the treatment of this form of leukemia but possibly for other cancers, as well.

Acute myeloid leukemia (AML) is the most common type of acute leukemia in adults. The average age of sufferers is 67, according to the American Cancer Society, and the disease has a dismal prognosis: the National Cancer Institute estimates a five-year survival rate of 22.8 percent.

Crowding out healthy cells
According to Ravindra Majeti, a hematologist at Stanford School of Medicine and an author on both studies, the problem is not necessarily the presence of the leukemia cells but how these cells interfere with blood-making. He notes that a person's bone marrow makes approximately 100 billion new blood cells daily. AML cells grow uncontrollably and over time crowd out the progenitor blood stem cells  made by the marrow. "So most of the real problems," including death, are a result of the body "no longer making blood," Majeti says, and not because of the cancer cells themselves.

Once made, these cancerous cells express a protein called CD47 on their surfaces. This protein, which is the same as that shown by normal blood stem cells, serves as a forged passport of sorts that it presents to leave the bone marrow. The security guards of the immune system, the macrophages, sense the CD47 protein and consequently pass the cell as "safe." Once through the checkpoint, the malignant cells can enter the blood stream and then can spread throughout the body.

Furthermore, AML cells produce more CD47 than healthy cells do to mask themselves, ensuring complete protection from the macrophages. And in previous work the group has shown that the more CD47 produced, the worse the prognosis.

Exploiting the CD47 protein shield
Although CD47 enables cancerous cells to evade the immune system, the protein could also lead to their undoing. In preliminary tests with mice infected with human AML, the Stanford group found an antibody that attaches to the CD47 protein, thereby masking this safety signal. As a result, macrophages attacked and removed the cells. The antibody does not go after normal blood stem cells as frequently, because those cells only express CD47 when they are on the move.

The researchers found a second method by which they could exploit the excessive amount of CD47. They were able to separate the cancerous cells from the noncancerous stem cells using a technique called fluorescence activated cell sorting. To do this, the group added a fluorophore, a molecule that emits fluorescent light, to the CD47 antibody. The fluorescing CD47 antibody then attached to any cells that have CD47 present. The cell sorter could separate cells with more fluorescence (the cancerous cells) from those with less fluorescence (noncancerous stem cells). The group hopes that this technique of isolating and removing the leukemia cells can help improve the results of autologous stem cell transplant—using a patient's own stem cells to restore the blood-producing system after chemotherapy or radiation.

Limitations of an idea
The cell sorting technique, however, has limits, observes Douglas Smith, an oncologist and leukemia expert at Johns Hopkins University's Sidney Kimmel Comprehensive Cancer Center. He explains that "there have been many different attempts to find drugs or antibodies to 'purge' the leukemia from a patient's own bone marrow or blood cells so that they can be used in transplants. However, these approaches are not always effective. Most of the time, the patients own leukemia returns following the transplant."

According to Majeti, "we do not fully understand why the bone marrow starts making these abnormal cells in the first place"; predicting the return of leukemia is difficult. And Smith notes that "introducing stem cells from a donor for such a transplant avoids the risk of reintroducing the patient's own leukemia and is one reason transplants from donors have a better outcome for most leukemias."

Still, given the current poor understanding and prognosis of the disease, Smith goes on to say that the technique is an advance in how we study leukemia and normal blood cells. David Ritchie and Mark Smyth, both at the Peter MacCallum Cancer Center in Australia, are also optimistic about the implications of these papers. In a preview published in Cell, they wrote that these results will "ultimately lead to improved outcomes of leukemia."

The Stanford group believes that their findings will have implications regarding other cancers, as well. Irving Weissman, the senior author on the studies, suspects that CD47-based stealth might also play a role in other cancers. "I believe this will be a general mechanism for cancer to evade macrophages," he says, noting a 1992 study led by Ian Campbell that showed 90 percent of ovarian cancer cells express CD47 on the cell surface. The Stanford team is investigating the possibility that many cancerous cells use the CD47 protein as a shield.  And if they're right, new weapons can be developed that may target a wide range of cancers.