Researchers at The Hospital for Sick Children, University Health Network and University of Toronto have shown that targeting SIRPA disrupts its interaction with CD47 on macrophages and could help eliminate acute myelogenous leukemia stem cells.1 The data provide validation for a SIRPA-Fc fusion protein being developed by Trillium Therapeutics Inc. that antagonizes the SIRPA-CD47 interaction-the biotech hopes to submit an IND in 2014.

CD47 is a membrane protein broadly expressed across tissues. It interacts with multiple proteins, including SIRPA (signal regulatory protein-a).

Normal cells express CD47 to protect themselves from phagocytosis-recent studies by researchers at Stanford University have shown that AML stem cells can do the same.2,3 The Stanford group also showed that an anti-CD47 mAb promoted macrophage-mediated phagocytosis of AML stem cells and prevented their engraftment in mouse models. AML stem cells are resistant to chemotherapy and have been linked to disease relapse.4

Although the Stanford researchers did suggest that the SIRPA-CD47 interaction mediated the phagocytosis-promoting effect, they were not able to provide direct evidence to show that this was indeed the case because the mAb they used may block CD47's interaction with additional known binding partners.5,6

The Toronto team decided to focus on SIRPA based on a study it published in 2007 showing that polymorphisms in the gene encoding SIRPA affected CD47 binding and the survival of engrafted normal human hematopoietic stem cells in mice.7

"We wanted to provide clarity as to what key molecule was interacting with CD47 to promote the phagocytosis of leukemia stem cells," said Jean Wang, an assistant professor in the Department of Medicine at the University of Toronto and a staff hematologist in the Division of Medical Oncology and Hematology in the Department of Medicine at the University Health Network.

Wang's team used a series of mice with polymorphisms in the gene encoding Sirpa and showed that disruption of Cd47's interaction with Sirpa is the key to promoting macrophage-mediated phagocytosis of AML stem cells.

The researchers confirmed their results with a SIRPA-Fc fusion protein. In primary AML cells, the SIRPA-Fc fusion protein increased macrophage-mediated phagocytosis of AML cells compared with the control IgG4-Fc fusion protein. In a mouse model of human AML, using the fusion protein to block the SIRPA-CD47 interaction decreased leukemic engraftment and growth compared with using a control IgG4-Fc fusion protein (see "Immune escape by leukemia stem cells and potential therapeutic strategies").

Importantly, disrupting the SIRPA-CD47 interaction using the SIRPA-Fc fusion protein enhanced macrophage-mediated killing of AML stem cells without enhancing the killing of healthy hematopoietic cells.

Results were published in The Journal of Experimental Medicine.

"What our work contributes that was not well established previously is that SIRPA is the key binding partner of CD47 and that this is the key interaction responsible for the survival and engraftment of leukemia stem cells," said corresponding author Wang.

"They also show that disrupting the SIRPA-CD47 interaction using a SIRPA-Fc fusion protein is well tolerated in animals and does not trigger phagocytosis of normal hematopoietic stem cells," added Robert Uger, VP of R&D at Trillium. "This suggests that there is a therapeutic window where we will be able to enhance macrophage-mediated killing of leukemia stem cells while sparing normal hematopoietic stem cells."

Ravi Majeti, an assistant professor of medicine in the Division of Hematology at Stanford, said induction of phagocytosis by the immune system is a new therapeutic mechanism to consider for AML.

"These studies take off from previous work with anti-CD47 targeting antibodies and really utilize the same mechanism of blocking the SIRPA-CD47 interaction. In that way, this study validates the mechanism and target," he said.

Trillium teams up

The Toronto group and Trillium have a SIRPA-Fc fusion protein in preclinical development to treat AML.

Uger said the Toronto group's 2007 study caught the company's initial attention. "We were very intrigued by the group's previous work exploring the role of the CD47-SIRPA interaction in the survival of normal hematopoietic stem cells," he told SciBX.

Indeed, in 2010 Trillium launched a preclinical program to develop a CD47-Fc fusion protein that agonizes and a SIRPA-Fc fusion protein that antagonizes the SIRPA-CD47 interaction and thus would help promote the engraftment of hematopoietic stem cell transplants.

However, Uger noted that the company is currently prioritizing the SIRPA-Fc program. He declined to disclose specific details.

"We wanted to move into the cancer stem cell space," he said. "The cancer stem cell story is best defined in AML, so it was a natural place for us to start."

Wang noted that the SIRPA-Fc fusion protein works by enhancing the immune system's ability to recognize leukemia stem cells, which sets it apart from targeted agents that disrupt specific regulatory pathways and targets on leukemia stem cells themselves.

"Patients could have leukemia subclones that have evolved and are dependent on different mechanisms to enhance growth and survival," she told SciBX. "Thus, targeted agents may not be able to eliminate subclones that don't depend on the particular targeted pathway."

In contrast, Wang noted that disrupting the CD47-SIRPA interaction acts on macrophages and enhances their ability to recognize all leukemia stem cells irrespective of heterogeneity in their intrinsic survival pathways.

Wang and Uger said Trillium's SIRPA-Fc fusion protein will most likely be used in combination with standard therapies for AML, such as chemotherapy, to help eliminate surviving leukemia stem cells and prevent disease relapse.

Wang also said the tissue distribution of SIRPA is more restricted than that of CD47, so disrupting the interaction from the SIRPA end could potentially have lower toxicity.

Uger said the company hopes to submit an IND in mid-to-late 2014.

The Hospital for Sick Children and University Health Network has filed a patent application covering compositions and methods for treating hematological cancers by targeting the SIRPA-CD47 interaction. Trillium licensed the IP in 2010.

Lou, K.-J. SciBX 5(38); doi:10.1038/scibx.2012.996
Published online Sept. 27, 2012


1.   Theocharides, A.P.A. et al. J. Exp. Med.; published online Sept. 3, 2012; doi:10.1084/jem.20120502
Contact: Jean C.Y. Wang, University Health Network, Toronto, Ontario, Canada

2.   Jaiswal, S. et al. Cell 138, 271-285 (2009)

3.   Majeti, R. et al. Cell 138, 286-299 (2009)

4.   Chan, W.-I. & Huntly, B.J.P. Semin. Oncol. 35, 326-335 (2008)

5.   Voit, S. et al. FEBS Lett. 544, 240-245 (2003)

6.   Stefanidakis, M. et al. Blood 112, 1280-1289 (2008)

7.   Takenaka, K. et al. Nat. Immunol. 8, 1313-1323 (2007)


The Hospital for Sick Children, Toronto, Ontario, Canada

Stanford University, Stanford, Calif.

Trillium Therapeutics Inc., Toronto, Ontario, Canada

University Health Network, Toronto, Ontario, Canada

University of Toronto, Toronto, Ontario, Canada