EHA Library - The official digital education library of European Hematology Association (EHA)

RED CELL EFFECTS OF THE ANTI-CD47 MONOCLONAL ANTIBODY HU5F9-G4 IN A PHASE I STUDY FOR RELAPSED OR PRIMARY REFRACTORY ACUTE MYELOID LEUKEMIA
Author(s): ,
Charlotte Brierley
Affiliations:
Dept of Haematology,Oxford University Hospitals NHS Foundation Trust,Oxford,United Kingdom
,
Julie Staves
Affiliations:
Dept of Haematology,Oxford University Hospitals NHS Foundation Trust,Oxford,United Kingdom
,
Corran Roberts
Affiliations:
Centre for Statistics in Medicine,Nuffield Department of Orthopaedics, Rheumatology and Musculoskeletal Sciences, University of Oxford,Oxford,United Kingdom
,
Hannah Johnson
Affiliations:
Oncology Clinical Trials Office,Department of Oncology, University of Oxford,Oxford,United Kingdom
,
Paresh Vyas
Affiliations:
MRC Molecular Haematology Unit,University of Oxford,Oxford,United Kingdom;Dept of Haematology,Oxford University Hospitals NHS Foundation Trust,Oxford,United Kingdom
,
Lawrence Goodnough
Affiliations:
Departments of Pathology and Medicine,Stanford University,Stanford,United States
Mike Murphy
Affiliations:
National Health Service Blood and Transplant,Oxford,United Kingdom;Dept of Haematology,Oxford University Hospitals NHS Foundation Trust,Oxford,United Kingdom
(Abstract release date: 05/17/18) EHA Library. Brierley C. 06/17/18; 214624; S1585
Charlotte Brierley
Charlotte Brierley
Contributions
Abstract

Abstract: S1585

Type: Oral Presentation

Presentation during EHA23: On Sunday, June 17, 2018 from 08:00 - 08:15

Location: Room A9

Background
CD47 is a physiological ‘don’t-eat-me’ signal to cells of the innate immune system, and its blockade results in phagocytic clearance of tumor cells in animal models. Downregulation of CD47 triggers clearance of ageing red blood cells (RBC).  Hu5F9-G4 (5F9) is a first-in-class antibody targeting CD47 being explored clinically in multiple tumors. In non-human primate studies 5F9 induced an on-target anemia by CD47 blockade and removal of circulating senescent RBC. 5F9 binding to RBCs may also confound blood type and antibody testing. 

Aims
Here we report the red cell effects, impact on blood bank testing and transfusion management in a Phase I trial of 5F9 in relapsed or primary refractory AML.

Methods
13 patients recruited at five centers in the U.K. were allocated to one of four escalating dose cohorts of 5F9. For each patient we collected serial data on hemoglobin (Hb), markers of hemolysis, transfusion requirements and any blood compatibility testing issues. On the day of each treatment, blood samples were taken pre-dose and post-dose (4 hrs +/- 30mins post infusion). Doses of 5F9 ranged from 0.1mg/kg to 15.0 mg/kg twice weekly. The cut-off for data collection was June 6 2017.

Results
All patients (13/13; 100%) experienced a decline in Hb between the pre- and post-dose blood samples. The median Hb decline following any dose of 5F9 across the course of the study was 1.2 g/dL (range 0.4 to 1.6). In 3 patients (23%), the Hb decreased post- 5F9 by 3 g/dL, all after the first dose. The maximum observed Hb drop in any patient was 5.2 g/dL. There was no consistent laboratory evidence of hemolysis as evaluated by bilirubin, LDH or reticulocyte counts. 13/13 patients (100%) developed evidence of RBC agglutination on blood smear examination.  No related clinically significant sequelae were observed. 12/13 (92%) of patients developed a positive direct antiglobulin test (DAT). DAT positivity was strongest after the first dose of 5F9 but remained detectable in 6/13 patients (48%) at end of treatment period. All patients (13/13; 100%) were transfused on-trial to a target Hb of 8-10g/dL. Transfusion requirements were higher on trial than pre-trial but this may be partially explained by a higher Hb threshold for transfusion after patients entered the trial. The median increment per unit of RBC was 1.0 g/dL which was maintained until the subsequent dose of 5F9. Discordant results from forward and reverse ABO typing were seen in 3/13 (23%), due to extra plasma reactivity in the reverse typing. 7/13 (54%) patients were found to have a positive antibody screen. In 6/13 (48%) patients this was a pan-agglutinin, and in 1/13 (8%) a known anti-E antibody. The median time to positive antibody screen was 2 days (range 2-114). Patients with a pan-agglutinin required matching of donor blood to the patient’s genotype to provide RBC for transfusion. All patients were safely transfused with no clinical complications. The antibody effect could not be abolished with DTT, chloroquine, papain, ficin, or trypsin.

Conclusion
AML patients treated with 5F9 may experience a decline in Hb and increased transfusion requirements. The likely mechanism of anemia in AML patients is clearance of aged RBCs by CD47 blockade, combined with a failure to mount a compensatory reticulocytosis. Type-specific RBC based on baseline ABO blood group may be required where ABO typing interference occurs.  Issues with compatibility testing because of newly positive antibody screens in ~50% of patients can be mitigated by pre-treatment geno/phenotype matching. Safe red cell transfusions were possible in all patients.

Session topic: 32. Transfusion medicine

Keyword(s): AML, Targeted therapy, transfusion

Abstract: S1585

Type: Oral Presentation

Presentation during EHA23: On Sunday, June 17, 2018 from 08:00 - 08:15

Location: Room A9

Background
CD47 is a physiological ‘don’t-eat-me’ signal to cells of the innate immune system, and its blockade results in phagocytic clearance of tumor cells in animal models. Downregulation of CD47 triggers clearance of ageing red blood cells (RBC).  Hu5F9-G4 (5F9) is a first-in-class antibody targeting CD47 being explored clinically in multiple tumors. In non-human primate studies 5F9 induced an on-target anemia by CD47 blockade and removal of circulating senescent RBC. 5F9 binding to RBCs may also confound blood type and antibody testing. 

Aims
Here we report the red cell effects, impact on blood bank testing and transfusion management in a Phase I trial of 5F9 in relapsed or primary refractory AML.

Methods
13 patients recruited at five centers in the U.K. were allocated to one of four escalating dose cohorts of 5F9. For each patient we collected serial data on hemoglobin (Hb), markers of hemolysis, transfusion requirements and any blood compatibility testing issues. On the day of each treatment, blood samples were taken pre-dose and post-dose (4 hrs +/- 30mins post infusion). Doses of 5F9 ranged from 0.1mg/kg to 15.0 mg/kg twice weekly. The cut-off for data collection was June 6 2017.

Results
All patients (13/13; 100%) experienced a decline in Hb between the pre- and post-dose blood samples. The median Hb decline following any dose of 5F9 across the course of the study was 1.2 g/dL (range 0.4 to 1.6). In 3 patients (23%), the Hb decreased post- 5F9 by 3 g/dL, all after the first dose. The maximum observed Hb drop in any patient was 5.2 g/dL. There was no consistent laboratory evidence of hemolysis as evaluated by bilirubin, LDH or reticulocyte counts. 13/13 patients (100%) developed evidence of RBC agglutination on blood smear examination.  No related clinically significant sequelae were observed. 12/13 (92%) of patients developed a positive direct antiglobulin test (DAT). DAT positivity was strongest after the first dose of 5F9 but remained detectable in 6/13 patients (48%) at end of treatment period. All patients (13/13; 100%) were transfused on-trial to a target Hb of 8-10g/dL. Transfusion requirements were higher on trial than pre-trial but this may be partially explained by a higher Hb threshold for transfusion after patients entered the trial. The median increment per unit of RBC was 1.0 g/dL which was maintained until the subsequent dose of 5F9. Discordant results from forward and reverse ABO typing were seen in 3/13 (23%), due to extra plasma reactivity in the reverse typing. 7/13 (54%) patients were found to have a positive antibody screen. In 6/13 (48%) patients this was a pan-agglutinin, and in 1/13 (8%) a known anti-E antibody. The median time to positive antibody screen was 2 days (range 2-114). Patients with a pan-agglutinin required matching of donor blood to the patient’s genotype to provide RBC for transfusion. All patients were safely transfused with no clinical complications. The antibody effect could not be abolished with DTT, chloroquine, papain, ficin, or trypsin.

Conclusion
AML patients treated with 5F9 may experience a decline in Hb and increased transfusion requirements. The likely mechanism of anemia in AML patients is clearance of aged RBCs by CD47 blockade, combined with a failure to mount a compensatory reticulocytosis. Type-specific RBC based on baseline ABO blood group may be required where ABO typing interference occurs.  Issues with compatibility testing because of newly positive antibody screens in ~50% of patients can be mitigated by pre-treatment geno/phenotype matching. Safe red cell transfusions were possible in all patients.

Session topic: 32. Transfusion medicine

Keyword(s): AML, Targeted therapy, transfusion

By clicking “Accept Terms & all Cookies” or by continuing to browse, you agree to the storing of third-party cookies on your device to enhance your user experience and agree to the user terms and conditions of this learning management system (LMS).

Cookie Settings
Accept Terms & all Cookies