Abstract
Introduction: Immune checkpoint blockade of CD47, a macrophage checkpoint protein that serves as a “don't-eat-me” signal on cancer cells, is known to enable tumor-cell phagocytosis, and has shown promising results in clinical trials in lymphomas. However, additional cytotoxic cell death mechanisms beyond phagocytosis have also been reported, potentially contributing to the overall anti-tumor activity. The latter cell death mechanisms have yet to be well-characterized, therefore warranting further investigation to comprehensively understand the anti-tumor mechanisms of CD47-blockade in lymphoid malignancies to facilitate identification of optimal drug partners for combination therapy.
Methodology: The fully humanized anti-CD47 monoclonal antibodies, SRF231, magrolimab, as well as the mouse monoclonal anti-CD47 antibody, B6H12, were evaluated in this study. Cell death mechanisms involving apoptosis, autophagy and necroptosis were analyzed using techniques such as BH3 profiling (Ryan, Biol Chem, 2016), Hoechst/Annexin V, flow cytometry, pharmacological and genetic (siRNA and CRISPR-Cas9 knockout) manipulations, Western blotting and immunohistochemistry. Cell lines including diffuse large B-cell lymphoma (DLBCL) (Ri-1, OCI-Ly1, TMD8, venetoclax-resistant OCI-Ly1-R), Burkitt's lymphoma (Raji), and T-acute lymphocytic leukemia (Jurkat) were used for in vitro analyses. Results generated from in vitro experiments were used to select promising drugs for treatment combination with CD47 blockade. Findings were validated using primary, peripheral blood-derived chronic lymphocytic leukemia (CLL) cells, purified human monocytes, T and B cells, treated ex vivo as well as in vivo in various lymphoid cell line-derived (CDX) and DLBCL patient-derived xenograft (PDX) mouse models.
Results: The anti-CD47 antibody SRF231 consistently killed tumor cells in lymphoid malignant cell lines, primary CLL cells, CDX and PDX models by activating necroptosis, while sparing other immune cells such as monocytes, T and B cells. The induction of necroptosis was confirmed by demonstrating increased phospho-activation of RIPK and downstream phospho-MLKL, which were abrogated by multiple pharmacological inhibitors of necroptosis and genetic silencing/knockout of CD47, RIPK and/or MLKL. We further ascertained that this necroptotic programmed cell death did not involve apoptotic- or autophagic-cell death, as caspase cleavage or LC3-flux were absent, respectively. Moreover, inhibiting apoptosis with Z-VAD-FMK or autophagy with BAPTA-AM did not rescue SRF231-induced cell death.
Given that necroptosis was the primary cell death mechanism, we utilized BH3 profiling technique to identify a potential combination partner to specifically induce apoptosis in our lymphoid malignant models, to potentially leverage an additional cell death pathway to enhance tumor cell death. BH3 profiling identified the BCL-2 inhibitor venetoclax as a potentially effective therapeutic partner for SRF231 against lymphoid malignant cells that are specifically and highly dependent on BCL-2 for survival (Survival rate: DMSO – 100%, (50nM) VEN – 71.39%, SRF231 – 57.01%, Combo – 19.53%, P-value < 0.0001). A treatment combination with SRF231 plus venetoclax in vivo completely eliminated tumor burden and prolonged progression-free survival in a BCL-2-dependent lymphoma xenograft model (Mice survival rate at day 70: Isotype control – 0%, VEN – 0%, SRF231 – 58.3%, Combo – 100%; P-value < 0.0001). Importantly, SRF231 was equally effective against non-BCL-2 dependent, venetoclax-resistant lymphoma models (PDX mice survival rate at day 200: Isotype control – 0%, SRF231 – 62.5%, P-value < 0.0001) .
Conclusion: Our study unravels a novel, critical non-canonical cell death mechanism of targeting CD47 via necroptosis. We also demonstrate the complementary and distinct cell death mechanisms of SRF231-induced necroptosis and venetoclax-induced apoptosis, a combination worthy of further study in the clinic specifically against BCL-2-dependent lymphoid malignancies.
