Leukemia cells evade immune system by mimicking normal cells, Stanford study shows

- By Krista Conger

STANFORD, Calif. — Human leukemia stem cells escape detection by co-opting a protective molecular badge used by normal blood stem cells to migrate safely within the body, according to a pair of studies by researchers at Stanford University Medical School.

“We call it the ‘Don’t eat me signal,’” said Ravindra Majeti, MD, PhD, assistant professor of hematology at the medical school and the co-first author of one of the studies, which focused on acute myeloid leukemia.

Patients whose cancer stem cells express higher levels of the molecule have a poorer prognosis than those whose cells express lower levels, and masking its presence makes the human cancer cells less deadly and more vulnerable to destruction when injected into mice. The results indicate that the molecule may serve both as a prognostic factor and a valuable therapeutic target for patients with the cancer.

“When we blocked this signal in mice with established human leukemia, the cancer cells were more easily removed by the body’s natural defenses,” Majeti said. The researchers are now moving ahead with plans to test a similar treatment in humans and have filed for a patent for the potential therapy.

Irving Weissman, MD, the Virginia & D.K. Ludwig Professor for Clinical Investigation in Cancer Research at the medical school, is the senior author of both studies, published together in the July 24 issue of the journal Cell. Majeti shares his first authorship with Mark Chao, an MD and PhD student in the cancer biology program at the medical school. Both Majeti and Weissman are members of Stanford’s Cancer Center.

Together, the researchers of the studies found that the molecule, CD47, protects the leukemia stem cells from macrophages — part of a roving cellular army tasked with finding and engulfing diseased or dying cells — by binding to a molecule on the macrophage’s surface. The interaction between the two proteins inhibits the macrophage’s killing instinct and allows the marauding cancer cells to escape unscathed.

The current research sprang from an earlier study in Weissman’s lab that showed CD47 expression was expressed at significantly higher levels in mouse leukemia cells. “At the time, we didn’t know what role CD47 played in leukemia,” said Siddhartha Jaiswal, an MD and PhD student at the medical school who is the first author of the second study. “But it was clearly important.”

It all fell into place, according to Weissman, when another group showed that red blood cells expressing CD47 were bypassed by macrophages, but those without CD47 were eaten. “Our discovery that leukemia cells in mice also expressed CD47 led us to propose that this signal might be appropriated by the leukemia stem cells as part of their leukemic progression,” said Weissman.

Jaiswal and his colleagues went on to show that although rapidly proliferating myeloid leukemia cells in human patients expressed about twice as much CD47 as their normal counterparts, CD47 levels were also temporarily elevated in healthy bone marrow stem cells migrating to organs like the spleen or the liver in response to infection or inflammation, or to bone marrow sites to replenish missing stem cells. A closer look determined why.

“These cells have to cross fields of macrophages guarding these tissues,” said Jaiswal. “Because the macrophages are also activated by the same stimuli, these elevated levels of CD47 protect the blood stem cells during this process.”

The macrophages’ destructive power is exemplified by the fact that mouse bone marrow stem cells lacking CD47 — and thus vulnerable to engulfment by the macrophages — are unable to rescue mice whose own immune systems have been destroyed. Conversely, a human myeloid leukemia line that normally expresses low levels of CD47 was able to cause leukemia in mice only after Jaiswal engineered it to express higher levels of the molecule. What’s more, when Jaiswal killed off the macrophages in mice, it allowed even the low-CD47 cell line to cause disease.

All told, the evidence points toward macrophages playing a starring, and unexpected, role in destroying cancer cells. Part of what’s called the innate immune system, macrophages and other protective cells cruise the body looking for all types of trouble. In contrast, the adaptive immune system mobilizes specialized T and B cells in response to specific activation signals.

“We hypothesized that, if elevated CD47 expression renders the leukemia stem cells invisible to the innate immune system, human patients with elevated levels would fare more poorly than those with lower levels,” said Majeti. “Furthermore, if we could prevent that interaction between the cancer cells and the macrophages, maybe the cells would get eaten.”

In fact, in three independent, previously published data sets comprising 664 human patients, those whose acute myeloid leukemia cells express higher levels of CD47 were about twice as likely to die from the cancer within a certain time period than those whose cancers expressed lower levels of the molecule: High CD47 expressors lived for a median of 9.1 months, while low CD47 expressors lived for a median of 22.1 months.

In further study of the high-CD47 expressing cancer stem cells in a culture dish, researchers added an antibody that blocks the interaction of the cancer cells with macrophages — essentially hiding the protective badge. This allowed the macrophages to engulf the cancer cells. A similar treatment in mice inhibited the human cancer cells’ ability to cause leukemia and even prolonged the survival of mice with previously established leukemias.

“This was the real kicker,” said Majeti. “These mice showed a profound clinical response.” Majeti and his colleagues are now planning to craft a similar antibody for use in human clinical trials.

Although the therapeutic promise is exciting, other avenues remain to be explored, say the researchers. For instance, Majeti and Chao showed that it is possible to separate leukemia stem cells from an individual patient into high and low CD47 groups. They found that the low-CD47 group contained normal blood stem cell progenitors.

“We’ve never before had the capability of reliably separating cancer cells from normal blood stem cells for bone marrow transplantation,” said Majeti, who pointed out that the normal stem cells could be used to rescue a patient whose immune system had been destroyed with lethal doses of cancer-killing radiation. If possible, this approach could theoretically eliminate the need for a matched donor by relying on the patient’s own cells. “It’s a dream come true for me as a physician-scientist to be researching something that is so applicable to patients,” said Majeti.

In addition to Majeti, Chao, Jaiswal and Weissman, other Stanford researchers involved in the studies include medicine instructor Ash Alizadeh, MD, PhD; graduate students Wendy Pang and Kenneth Gibbs, Jr.; former post-doctoral fellow Catriona Jamieson, MD, PhD; and pathology instructor Chris Park, MD. The research was funded by the National Institutes of Health, the Leukemia Society, the Ludwig Institute and the Smith Family Fund.

About Stanford Medicine

Stanford Medicine is an integrated academic health system comprising the Stanford School of Medicine and adult and pediatric health care delivery systems. Together, they harness the full potential of biomedicine through collaborative research, education and clinical care for patients. For more information, please visit med.stanford.edu.