-Author name in bold denotes the presenting author
-Asterisk * with author name denotes a Non-ASH member
Clinically Relevant Abstract denotes an abstract that is clinically relevant.

PhD Trainee denotes that this is a recommended PHD Trainee Session.

Ticketed Session denotes that this is a ticketed session.

2834 Co-Targeting of CD38 and CD47 in T Cell Acute Lymphoblastic Leukemia

Program: Oral and Poster Abstracts
Session: 614. Acute Lymphoblastic Leukemia: Therapy, excluding Transplantation: Poster III
Hematology Disease Topics & Pathways:
Leukemia, ALL, Biological, antibodies, Diseases, Therapies, immunotherapy, Lymphoid Malignancies
Monday, December 7, 2020, 7:00 AM-3:30 PM

Fotini Vogiatzi, PhD1*, Kristina Müller1*, Dorothee Winterberg2*, Thies Rösner, PhD3*, Lennart Lenk, PhD4*, Gunnar Cario, MD, PhD5*, Martin Schrappe, MD PhD1, Andreas E. Kulozik, MD, PhD6*, Beat Bornhauser7*, Jean-Pierre Bourquin, MD, PhD8, Thomas Valerius, MD3, Matthias Peipp, PhD3*, Christian Kellner, PhD9* and Denis Martin Schewe, MD10*

1Pediatric Hematology/Oncology, ALL-BFM Study Group, Christian Albrechts University Kiel and University Hospital Schleswig-Holstein, Campus Kiel, Germany, Kiel, Germany
2Pediatric Hematology/Oncology, ALL-BFM Study Group, Christian Albrechts University Kiel and University Hospital Schleswig-Holstein, Campus Kiel, Germany, Kiel, DEU
3Division of Stem Cell Transplantation and Immunotherapy, Department of Medicine II, Christian Albrechts University Kiel and University Hospital Schleswig-Holstein, Campus Kiel, Germany, Kiel, Germany
4UKSH Kiel, Kiel, Schleswig Holstein, Germany
5Pediatric Hematology/Oncology, ALL-BFM Study Group, Christian Albrechts University Kiel and University Hospital Schleswig-Holstein, Campus Kiel, Kiel, Schleswig-Holstein, Germany
6Department of Pediatric Oncology, Hematology and Immunology, University of Heidelberg and Hopp Children’s Cancer Center at Nationales Centrum for diseases Heidelberg, Heidelberg, Germany, Heidelberg, Germany
7Division of Pediatric Oncology, and Children Research Center, University Children´s Hospital, Zurich, Switzerland, Zurich, Switzerland
8Universitaets-Kinderspital Zurich, Zurich, Switzerland
9Division of Transfusion Medicine, Cell Therapeutics and Hemostaseology, University Hospital, LMU Munich, Munich, Germany, Munich, Germany
10Pediatric Hematology/Oncology, Kiel, Germany

Antibody application is a promising therapy in hematological malignancies including acute lymphoblastic leukemia (ALL). Unlike for B-cell precursor (BCP-ALL), immunotherapeutic interventions in T-cell ALL (T-ALL) are practically non-existent. Most T-ALL patient samples show substantial surface expression of CD38. Moreover, mice bearing T-ALL patient-derived xenograft (PDX) samples treated with daratumumab (Dara) monotherapy displayed prolonged survival and MRD-negativity in 50% of cases as opposed to animals treated with chemotherapy (Vogiatzi et al., Blood, 2019). Besides CD38, elevated surface expression of CD47 has been described in T-ALL (Chao et al., 2011). CD47 acts as a “don’t-eat-me“ signal protecting cancer cells from macrophage-dependent phagocytosis. In this study, we explored the efficacy of Dara and a CD47 blocking antibody (Fc-modified version of Hu5F9-G4, termed Hu5F9-IgG2σ) alone or in combination in T-ALL.

In vitro phagocytosis assays with M-CSF derived M0 (CD64+, CD80-, CD163+) macrophages from healthy donors were evaluated in T-ALL cell lines. Cells labelled with a pH-dependent dye became fluorescent upon engulfment by macrophages and formation of the acidic phagolysosome, which was automatically measured as an absolute cell count. The highest count was set as 100% in each experiment and all other values expressed as percentages relative to that. Treatment of MOLT-13 cells with Dara or Hu5F9-IgG2σ increased median phagocytosis from 5% in control to 23% and 27%, respectively (p=0.008 for both, Figure 1A). However, combined treatment resulted in a maximal increase of median phagocytosis in MOLT-13 cells (p=0.004 compared to Dara and p=0.008 compared to Hu5F9-IgG2σ, Figure 1A). Similar results were obtained in HSB-2 and P-12 cells (p=0.037/0.004 for HSB-2, p=0.011/0.001 for P-12). In addition, phagocytosis was determined in 12 PDX samples from random T-ALL patients. Hu5F9-IgG2σ alone showed a modest increase in median phagocytosis compared to control (19% vs. 6%, p<0.001) and Dara increased phagocytosis to 43% (p<0.001). Combination therapy, however, led to a maximal rise of median phagocytosis in all samples (p<0.001 compared to all others). Phagocytosis was also determined in 16 relapsed/refractory (r/r) T-ALL-PDX samples, in which a maximal median phagocytosis upon combination of Dara and Hu5F9-IgG2σ was also observed (2% in control, 37% in Dara and 14% in Hu5F9-IgG2σ, p<0.001/p=0.02/p=0.02, respectively). The extent of phagocytosis correlated with CD38 and CD47 surface expression on T-ALL cells and, in part, also with expression of signal regulatory protein α (SIRPα) on macrophages, the interacting partner of CD47. A maximal increase in phagocytosis was also observed when applying Hu5F9-IgG2σ in 1/10 of the concentration of Dara, confirming its high efficacy.

The effects of the Dara/ Hu5F9-IgG2σ combination were also examined in vivo in phase II-like preclinical trials containing different patients in one experiment. Six random T-ALL-PDX samples were injected into NOD.Cg-Prkdcscid Il2rgtm1Wjl/SzJ (NSG) mice and subjected to therapy with Dara and Hu5F9-IgG2σ alone or in combination. In order to mimic minimal residual disease (MRD), antibody therapy was started one day post injection. Mice subjected to Dara displayed prolonged survival in 4/6 (67%) animals compared to control (p=0.007), whereas animals treated with Hu5F9-IgG2σ showed significantly prolonged survival in all cases (p=0.007, Figure 1B). Interestingly, mice not responding to Dara benefitted from Hu5F9-IgG2σ monotherapy and the combination had no further survival advantage in this MRD model. In a second approach, eight r/r T-ALL PDX samples were used, and the same therapy was started upon overt leukemia (1% human blasts in peripheral blood). Compared to control, mice treated with Dara in this overt leukemia and r/r setting showed no survival benefit, while animals subjected to Hu5F9-IgG2σ showed a trend towards a prolonged survival (5/8 cases, 63%, p=0.08). Most importantly, the combination of both agents resulted in a statistically significant survival prolongation in 7/8 cases (88%, p=0.010, Figure 1C).

Altogether, we show that combining Dara with CD47 blockade increases phagocytosis in vitro in three T-ALL cell lines and 28 PDX samples. Our in vivo data suggest that this combination is a highly promising therapeutic strategy, especially in the relapsed/refractory setting.

Disclosures: Cario: Jazz Pharmaceuticals: Consultancy, Other: travel support; Novartis: Consultancy, Other: travel support. Kulozik: bluebird bio, Inc.: Consultancy, Honoraria; Novartis: Consultancy, Honoraria. Schewe: Sobi: Other: was an advisory board member; Bayer: Other: was an advisory board member; Jazz: Other: was an advisory board member; OSE pharmaceuticals: Research Funding.

*signifies non-member of ASH