Preclinical Evidence for the Efficacy of CD79b Immunotherapy in B-cell Precursor Acute Lymphoblastic Leukemia

1Department of Pediatrics I, ALL-BFM Study Group, Christian-Albrechts University Kiel and University Medical Center Schleswig-Holstein, Kiel, Germany 2Department of Medicine II, University Hospital Schleswig-Holstein, Kiel, Germany 3Department of Pathology, Section of Pediatric Pathology, University Hospital Bonn, Germany 4Department of Pathology, Hematopathology Section and Lymph Node Registry, Christian-Albrechts University Kiel and University Medical Center SchleswigHolstein, Kiel, Germany 5Genentech Research and Early Development, San Francisco, CA, USA 6Institute of Molecular and Clinical Immunology, Medical Faculty, Otto-VonGuericke University Magdeburg, Germany 7Molecular and Experimental Surgery, Clinic for General, Visceral, Vascular, and Transplant Surgery, Medical Faculty, University Hospital Magdeburg, Germany 8Department of Medicine II, Division of Antibody-Based Immunotherapy, Christian-Albrechts University Kiel and University Medical Center SchleswigHolstein, Kiel, Germany 9Department of Medicine II, Section for Stem Cell Transplantation and Immunotherapy, Christian-Albrechts University Kiel and University Medical Center Schleswig-Holstein, Kiel, Germany 10Institute of Immunology, Ulm University Medical Center, Ulm, Germany 11Department of Pediatrics, Otto-von-Guericke-University, Magdeburg, Germany LL and DW have contributed equally to this work. Supplemental digital content is available for this article. Copyright © 2022 the Author(s). Published by Wolters Kluwer Health, Inc. on behalf of the European Hematology Association. This is an open access article distributed under the Creative Commons Attribution License 4.0 (CCBY), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. HemaSphere (2022) 6:8(e754). http://dx.doi.org/10.1097/HS9.0000000000000754. Received: April 20, 2022 / Accepted: June 16, 2022 Preclinical Evidence for the Efficacy of CD79b Immunotherapy in B-cell Precursor Acute Lymphoblastic Leukemia

Correspondence: Denis M. Schewe (denis.schewe@med.ovgu.de). P atients with B-cell precursor acute lymphoblastic leukemia (BCP-ALL) have a favorable prognosis. However, current treatment protocols are based on intensive cytotoxic chemotherapy and therapy options are limited when patients relapse. 1 Hence, there is an urgent need for novel immunotherapy approaches.
A potential novel target could be the pre-B-cell receptor (pre-BCR) signaling complex for which an integral role in B-cell malignancies has been proposed. 2 The pre-BCR consists of the µ heavy chain (µHC), the surrogate light chain (VpreB and Lambda 5), and a signaling heterodimer composed of CD79a (Igα) and CD79b (Igβ). Activation of the pre-BCR induces downstream signaling via phosphorylation of tyrosine residues within the immunoreceptor tyrosine-based activation motifs (ITAMs) of the cytoplasmic tails of CD79a/CD79b. 3 Due to its high abundance on mature B cells, the BCR complex has become an important target for diffuse large B-cell lymphoma (DLBCL) treatment and the CD79b antibody drug conjugate (ADC) Polatuzumab Vedotin (PolVed) has shown therapeutic efficacy in DLBCL-frontline treatment. 4 Nevertheless, the CD79a/CD79b heterodimer is already expressed at the pro-B-cell stage before a productive immunoglobulin gene rearrangement is accomplished, even without associated µHC. 5 Moreover, we recently reported that the pre-BCR signaling unit CD79a is crucial for BCP-ALL engraftment in vivo, particularly in the central nervous system (CNS), in BCR-ABL1 + and E2A-PBX1 + patient-derived xenograft (PDX) models. 6 We therefore hypothesized that CD79b may also serve as a therapeutic target in BCP-ALL.
Here, we show that surface (s)CD79b is expressed in different subgroups of pediatric BCP-ALL patients and that targeting CD79b with a monoclonal antibody reduced CNS involvement of sCD79b positive (sCD79b + ) PDX samples in vivo. Moreover, the CD79b-ADC PolVed significantly diminished overall leukemia burden and prolonged mouse survival in sCD79b + BCP-ALL-PDX models of different cytogenetic backgrounds.
First, to test whether CD79b is important for leukemic engraftment, we applied a murine/murine transplantation model as described previously. 6 Murine precursor B cells harboring a truncated cytoplasmic tail of CD79b, which results in the loss of the ITAM and therefore functional CD79b 7 (referred to as CD79b-ITAM-KO) were malignantly transformed by stable transduction with a BCR-ABL1-construct (Suppl. Figure S1A-B). Transformed CD79b-ITAM-KO cells showed a similar proliferation pattern as control (Ctr) cells in vitro (Suppl. Figure S1C). However, upon transplantation into NOD scid gamma (NSG) mice, Ctr cells exposed profound leukemia engraftment in the spleen (Sp; P = 0.0079), bone marrow (BM, P = 0.0079), and CNS (P = 0.0079) leading to sacrifice of all Ctr animals after 25 days while animals injected with CD79b-ITAM-KO cells did not show signs of leukemia at that timepoint ( Figure 1A-C, Suppl. Figure S1D). An additional group of mice transplanted with CD79b-ITAM-KO cells was left for survival analysis and mice of this group were still free of leukemia upon termination of the experiment after 162 days ( Figure 1C, Suppl. Figure S1E-F) indicating that CD79b is required for leukemia engraftment in vivo.
Due to the lack of BCP-ALL engraftment of transformed CD79b-ITAM-KO cells, we hypothesized that blocking CD79b with a monoclonal antibody reduces BCP-ALL engraftment in vivo. To test this hypothesis, we applied an unconjugated CD79b-IgG1-antibody (clone SN8, anti-CD79b) on NSG mice bearing either E2A-PBX1 + or BCR-ABL1 + PDX samples (Suppl . Table S1/ PDX1-2) with high sCD79b expression (53.6% sCD79b + cells and 25.9% sCD79b + cells, respectively, Suppl. Figure S3, Suppl. Figure S4A-B). Animals were injected with 1 × 10 5 PDX cells and anti-CD79b-treatment was initiated 1 day post-injection ("minimal residual disease [MRD] model" 8,10,11 ). All animals were sacrificed when the first mouse showed clinical signs of leukemia. Anti-CD79b-immunotherapy resulted in a small reduction of leukemia burden in the Sp and BM of E2A-PBX1 + PDX mice and a significant reduction of Sp and BM engraftment in BCR-ABL + PDX mice ( Figure 1E and F, Suppl. Figure S4C-F). Of note, anti-CD79b-treatment promoted a significant reduction of CNS involvement in both PDX models (P = 0.0302 and P = 0.0098, respectively; Figure 1G-I) indicating that CD79b blockade impacts the engraftment of BCP-ALL cells in vivo and survival of BCP-ALL cells in the CNS.
Next, we investigated if CD79b-immunotherapy using a CD79b-ADC would outperform the efficacy of CD79b blockade. Hence, we performed the same in vivo experiment applying PolVed therapy in our sCD79b + PDX-MRD models (Suppl. Table S1/PDX1-2). Indeed, PolVed therapy led to a profound anti-leukemic effect in both PDX models showing significantly reduced Sp sizes as well as blast numbers in Sp and BM as compared with Ctr animals (P < 0.0001, respectively; Figure 2A and B; Suppl. Figure S5A-B). Accordingly, all PolVed-treated PDX mice were CNS negative upon sacrifice of Ctr animals (P = 0.0079, respectively; Figure 2C and D). Analysis of further PolVed-treated animals (n = 5), which were left for survival analysis, showed a significant survival prolongation under PolVed therapy in both PDX models (median overall survival [MOS] 106 d versus not definable; P = 0.0027 and 76 versus 148 d; P = 0.0027; Figure 2E and F). Of note, 4/5 PolVed-treated E2A-PBX1 + PDX mice were free of BCP-ALL-PDX cells upon termination of the experiment after 236 days ( Figure 2E, Suppl. Figure S5C).
To further validate the target-specificity of PolVed in sCD79b + BCP-ALL cells, we compared the efficacy of PolVed to that of the CD30-ADC Brentuximab Vedotin (BreVed) in vivo in the with BCP-ALL-PDX cells from an E2A-PBX1 + and a BCR-ABL + patient and treated with the CD79b-ADC PolVed (1 mg/ kg, n = 10) or a Ctr vehicle (n = 5) starting the day after injection, modeling an MRD situation (intravenous treatment on day +1, +3, +7, +14 and every 14 d thereafter as described previously 8 ). Five animals, respectively, were sacrificed when the first mouse showed signs of overt leukemia (such as ataxia, splenomegaly, weight loss, or >70% leukemic cells in the PB; all Ctr animals had developed overt leukemia at this time point). One group of mice treated with PolVed (n = 5) was maintained for survival analysis. (A and B), Volumes of extracted spleens (indicator for leukemic engraftment) were measured, unpaired 2-sided t test. (C and D), CNS infiltration was assessed by semi-quantitative scoring as described previously, 6,8 Fisher exact test. (E and F), Therapy-associated differences in the survival of NSG mice bearing E2A-PBX1 + or BCR-ABL + BCP-ALL cells were determined using Kaplan-Meier log-rank statistics. The experiment was terminated after 236 d and 4/5 BCR-ABL + PDX mice treated with PolVed were sacrificed without showing signs overt leukemia. (G-I), A phase 2-like PDX study was performed using sCD79b-positive (≥10% sCD79b + cells, n = 4), and CD79b-negative (<10% sCD79b + cells, n = 8) PDX samples from different cytogenetic subgroups (5xE2A-PBX1 + , 3xBCR-ABL + , 2xMLLr, 1xE2A-HLF + , and 1xETV6-NTRK3 + ). Two NSG mice per patient were injected with PDX cells, randomly assigned into treatment groups and PolVed therapy was initiated upon detection of 1% PDX cells in the PB, modeling an overt leukemia situation. (G), Blood of both, Ctr and PolVed treated animals bearing the same PDX sample was withdrawn when one of the 2 PDX mice showed signs of overt leukemia and the number of hCD45 + /hCD19 + / mCD45cells in the PB was measured via flow cytometry. The waterfall plot shows the difference in PB blasts between respective Ctr and PolVed treated mice normalized to the maximum blast reduction (sorted from weakest therapy response to highest therapy response). Animals not showing clinical signs of overt leukemia or >70% PB blasts at this timepoint received further treatment until reaching termination criteria. Therapy-associated differences in the survival of NSG mice bearing (H) sCD79band (I) sCD79b + PDX cells were determined using Kaplan-Meier log-rank statistics. ADC = antibody drug conjugate; BCP-ALL = B-cell precursor acute lymphoblastic leukemia; BCP-ALL-PDX = B-cell precursor acute lymphoblastic leukemia-patient-derived xenograft; CNS = central nervous system; Ctr = control; hCD45+/hCD19+/ mCD45-= human (h)CD45+hCD19+(murine) mCD45-; MLLr = MLL rearranged; MRD = minimal residual disease; ns = not significant; NSG = NOD scid gamma; PB = peripheral blood; PDX = patient-derived xenograft; PolVed = Polatuzumab Vedotin.
E2A-PBX1 + sCD79b + /CD30 -PDX model (Suppl. Figure S6A, Suppl. Table S1/PDX1). PolVed had anti-leukemic efficacy and BreVed treatment resulted in Sp sizes and blast counts in Sp and BM comparable with that of Ctr animals suggesting that PolVed kills BCP-ALL cells in a target-specific manner (Suppl. Figure  S6B-C).
Antibody-based immunotherapies such as Blinatumomab have become an important tool in BCP-ALL treatment. Yet, the observation of tumor immune-escape via downregulation of the target-antigen, for example, CD19 motivates the identification of novel immunotherapy targets. 13 We show the presence of CD79b on the surface of diagnostic patient samples of different BCP-ALL subgroups. This is particularly interesting as previous reports suggested that only certain cytogenetic subgroups such as E2A-PBX1 + BCP-ALL are considered as pre-BCR positive, whereas most BCP-ALL cases, including BCR-ABL + BCP-ALL do not express µHC and therefore a functional pre-BCR on the cell surface. 14 Our data promote the view that CD79b is expressed on the surface of BCP-ALL cells irrespective of a fully assembled pre-BCR signaling complex as previously hypothesized. 6 This is further supported by the recent finding that the pre-BCR surrogate light chain component VpreB was detected in subpopulations of BCP-ALL patient samples regardless of cytogenetic subgroups. 15 Yet, the role of CD79b and VpreB (and other BCR components) may differ markedly in BCP-ALL. Unlike VpreB, CD79b harbors an ITAM in the cytoplasmic domain by which CD79b on the cell surface may promote downstream signaling, irrespective of a fully arranged pre-BCR complex, thereby enhancing the survival and proliferation of BCP-ALL cells. Accordingly, in our model CD79b deletion had a direct effect on ALL propagation in vivo, indicating a functional role in ALL-pathogenesis. Since CD79b immunotherapy has already entered clinical routine in other B-cell malignancies, 4 PolVed therapy may represent an interesting treatment alternative for BCP-ALL, potentially also in relapsed/refractory disease. To this end, PolVed treatment was also effective in BCR-ABL + , MLLr and E2A-HLF + PDX samples, which are considered highrisk subgroups. The prospective measurement of sCD79b+ in newly diagnosed and (CD19 -) relapsed BCP-ALL patients may help to identify patients who could benefit from CD79b immunotherapy. Moreover, an important step before clinical transition will be to preclinically test the efficacy and tolerability of PolVed in combination with standard-of-care treatments using PDX models. 10,11 In this respect, the efficacy of CD79b immunotherapy could be tested in comparison or combination with small molecule inhibitors targeting the pre-BCR signaling pathway. 14 Overall, gaining a better understanding of the role of the various components of the pre-BCR in leukemia development and relapse may improve diagnostic and therapeutic options in BCP-ALL.