# CD47 / SIRPα (the 'don't eat me' signal)

Source: https://onco.cc/pathways/cd47-sirpa/  
OnCo record `cd47-sirpa` (Pathway). Data CC BY-NC 4.0, attribute "Data from OnCo (onco.cc)"; commercial use needs a licence.

## TL;DR

Macrophages eat cells that look wrong unless the cell shows CD47, a 'don't eat me' badge that binds SIRP-alpha. Leukaemias, myelodysplastic syndromes, lymphomas and carcinomas overproduce CD47 to escape being eaten; antibodies that cover the badge work in the lab, but the leading drug, magrolimab, failed in blood cancers because red cells wear the same badge and lose it too.

## Summary

CD47 on a cell binds SIRPα on macrophages and dendritic cells; SIRPα's ITIM motifs recruit SHP-1 and SHP-2, which stop the myosin-driven engulfment programme. Healthy cells, including red cells and platelets, rely on it; tumours (AML, MDS, lymphoma, many carcinomas) express high CD47 to defeat the 'eat me' signals (calreticulin, phosphatidylserine, antibody Fc bound to the cell) that would otherwise trigger phagocytosis. Blocking CD47 or SIRPα therefore synergises with opsonising antibodies (rituximab) and with agents that expose calreticulin (azacitidine). Magrolimab, the first anti-CD47 antibody, showed early promise with azacitidine in MDS and AML but the phase 3 ENHANCE trials were stopped in 2023-2024 for futility and excess deaths, and the programme was discontinued; on-target anaemia (red cells express CD47) limits every agent in the class. Newer designs try to reduce red-cell binding (low-affinity or Fc-silent antibodies, SIRPα-Fc fusion proteins such as evorpacept, CD47 × tumour-antigen bispecifics) and the checkpoint is being explored in solid tumours with anti-EGFR and anti-HER2 antibodies.

## Fields

- Kind: Pathway
- Last checked: 2026-09-04
- Analogy: Macrophages are security guards who remove anyone without a staff badge. Cancer cells forge the badge (CD47) so the guards wave them through. Anti-CD47 drugs cover the forged badges, but every red blood cell carries a genuine badge too, so covering them all makes the guards remove red cells as well: anaemia.
- Interventions: Anti-CD47 magrolimab with azacitidine: phase 3 ENHANCE, ENHANCE-2 and ENHANCE-3 stopped for futility or harm; programme discontinued (2024); SIRPα-Fc fusions (evorpacept) and Fc-silent or low-affinity anti-CD47 antibodies designed to spare red cells; CD47 × CD19 or CD20 bispecifics to confine blockade to B-cell tumours; Combination with opsonising antibodies (rituximab, cetuximab, trastuzumab) to supply the 'eat me' signal

## Sources

- Wikipedia: https://en.wikipedia.org/wiki/CD47
- Wikipedia: https://en.wikipedia.org/wiki/CD47

## Connected records

- cancers: [Acute myeloid leukaemia](https://onco.cc/cancers/aml/), [Diffuse large B-cell lymphoma](https://onco.cc/cancers/dlbcl/), [Myelodysplastic syndromes / neoplasms (MDS)](https://onco.cc/cancers/mds/), [Non-Hodgkin lymphoma (all types)](https://onco.cc/cancers/non-hodgkin-lymphoma/)
- targets: [CD47](https://onco.cc/targets/cd47/), [PTPN6](https://onco.cc/targets/ptpn6/)
- drugs: [Magrolimab](https://onco.cc/drugs/magrolimab/)
- pathways: [Antigen presentation & immune editing](https://onco.cc/pathways/antigen-presentation-immunoediting/), [Tumour microenvironment (TME)](https://onco.cc/pathways/tumor-microenvironment/)
- terms: [What it costs to aim at a lineage antigen](https://onco.cc/terms/lymphoma-bio-lineage-antigen-cost/)

---
JSON: https://onco.cc/api/v1/entities/cd47-sirpa.json