Sig15 promotes an immunosuppressive tumor microenvironment in B-ALL. A–M, Unirradiated WT C57BL6 mice were injected via tail vein with 5 × 105 Cas9-only (n = 8 from two independent experiments) or Sig15 KO (n = 8) leukemia. Bone marrow was harvested 7 days later for highly dimensional flow cytometry of non-leukemia bone marrow populations. Healthy, leukemia-naïve mice (Naive) were included as controls (n = 6). CD3+ T cells (A), NK cells (B), classical dendritic cells (cDC; C), and neutrophils (D) were variably present at higher totals in the bone marrow of Sig15 KO recipient mice compared with control leukemia. E, Representative contour plots of T-cell populations. Proportions of CD8+ (F) but not CD4+ (G) T cells were higher in Sig15 KO recipients than control leukemia, increasing the CD8+/CD4+ T-cell ratio (H) to levels comparable with Naive mice. I, Degranulated CD8+ T cells were higher in Sig15 KO recipient mice, as were SLEC (J) and MPEC (K) CD8+ T-cell populations. Early memory-like populations consistent with central (L) and effector memory (M) CD8+ T-cell populations were also higher in Sig15 KO recipients. N–P, From the same experiments, bone marrow supernatant from control and Sig15 KO leukemia recipient mice was analyzed via 44-plex cytokine/chemokine assay. Levels of IL6 (N), LIF (O), and IL5 (P) were all significantly lower in the bone marrow of Sig15 KO recipients as compared with control leukemia (*, P < 0.05; **, P < 0.01; ***, P < 0.001; ****, P < 0.0001). Sig15 expression on B-ALL cells suppresses CD8+ T-cell expansion and activation and contributes overall to formation of a more proleukemia bone marrow microenvironment.

Siglec-15 (Sig15) has been implicated as an immune checkpoint expressed in solid tumor-infiltrating macrophages and is being targeted in clinical trials with mAbs to normalize the tumor immune microenvironment and stimulate antitumor immunity. However, the role of Sig15 in hematologic malignancies remains undefined. Sig15 mRNA and protein expression levels in hematologic malignancies were determined from publicly available databases, cell lines, and primary patient samples. Human B-cell acute lymphoblastic leukemia (B-ALL) cell lines were used to identify signaling pathways involved in the regulation of Sig15 expression. Secreted/soluble Sig15 and cytokine levels were measured from the plasma of children with leukemia and healthy controls. Knockdown and knockout of Siglec15 in a murine model of B-ALL was used to evaluate the effect of leukemia-derived Sig15 on the immune response to leukemia. We observed pathologic overexpression of Sig15 in a variety of hematologic malignancies, including primary B-ALL samples. This overexpression was driven by NFκB activation, which also increased the surface localization of Sig15. Secreted/soluble Sig15 was found to circulate at elevated levels in the plasma of children with B-ALL and correlated with an immune-suppressive cytokine milieu. Genetic inhibition of Sig15 in murine B-ALL promoted clearance of the leukemia by the immune system and a marked reversal of the immune-privileged leukemia bone marrow niche, including expanded early effector CD8+ T cells and reduction of immunosuppressive cytokines. Thus, Sig15 is a novel, potent immunosuppressive molecule active in leukemia that may be targeted therapeutically to activate T lymphocytes against leukemia cells. Significance: We demonstrate that Sig15 is overexpressed in hematologic malignancies driven by NFκB, is required for immune evasion in a mouse model of leukemia, and, for the first time, that it circulates at high levels in the plasma of children with leukemia.

Sig15 is required for immune evasion by murine B-ALL.

B-ALL–expressed Sig15 has dynamic subcellular localization. A, Flow cytometry of human B-ALL cell lines probed for SIG15 in both permeabilized and unpermeabilized (Surface) cells. Dotted line marks mean fluorescence intensity (MFI) of healthy PBMC signal (data representative of three independent experiments). Leukemia and PBMCs had modest surface expression of SIG15, while leukemia cells expressed higher intracellular SIG15 compared with PBMCs. SIG15 was probed using the NP159 mAb (NextCure). B and C, Immunofluorescence of REH cells staining for DAPI, Sig15, and Golgin-97. B, Imaging depicts limited surface localization of Sig15 and strong colocalization with the Golgi apparatus and TGN in unstimulated REH cells. C, Representative images depict REH cells treated for 24 hours with DMSO (vehicle control) or 81 nmol/L PMA and 1.3 μmol/L ionomycin. Stimulated REH cells show decreased colocalization of Sig15 signal with Golgin-97. SIG15 was probed using the Invitrogen polyclonal antibody. D, Flow cytometry of unpermeabilized REH cells treated with PMA with or without BOT64 for 6 hours, demonstrating that NFkB activation stimulates Sig15 localization at the cell surface (data representative of three independent experiments; dotted line represents MFI of isotype staining). SIG15 was probed using the NP159 mAb (NextCure) conjugated to Alexa Fluor 647.

Sig15 expression is regulated by NFκB activation in B-ALL. A, Western blot analysis shows upregulated SIG15 expression in REH cells stimulated for 24 hours with a PMA/ionomycin stimulation (81 nmol/L PMA and 1.3 μmol/L ionomycin) but not with 30 ng/mL of recombinant human MCSF and/or IL10. Densitometric analysis of SIG15 is quantified above. SIG15 was probed using the Invitrogen polyclonal antibody. B, qRT-PCR from REH cells treated with PMA (81 nmol/L) with or without pan-PKC inhibitor Gö 6983 (5 μmol/L) for 24 hours, showing that PKC inhibition abrogates SIG15 induction by PMA. (****, P < 0.0001). C, qRT-PCR from REH cells stimulated with PMA (81 nmol/L) and/or treated with calcineurin inhibitor CSA (1 μmol/L) or IKK-2 inhibitor BOT64 (10 μmol/L) for 24 hours, demonstrating that NFκB inhibition but not calcineurin inhibition abrogates PMA-induced SIG15 transcription (****, P < 0.0001). D, Western blot analysis shows SIG15 expression in REH cells following 24 hours of stimulation with 81 nmol/L PMA alone or in combination with 10 μmol/L of the IKK-2 inhibitor BOT64. NFκB inhibition attenuates PMA-induced upregulation of SIG15. SIG15 was probed using the 1F7 mAb (NextCure). E,SIG15 expression from the B-ALL TARGET database correlated with protein kinase C β (PRKCB; r = 0.35; P < 0.0001, Pearson correlation), calcineurin catalytic subunit alpha (PPP3CA; r = 0.42; P < 0.0001), and IL1β (IL1B; r = 0.33; P < 0.0001, Pearson correlation). F, Western blot analysis shows increased SIG15 protein in REH cells stimulated for 24 hours with increasing doses of recombinant human CD40L. Densitometric analysis of SIG15 is quantified above. SIG15 was probed using the 1F7 mAb (NextCure).

Sig15 ablation in murine B-ALL promotes an adaptive immune response against leukemia.

Pathological overexpression of Sig15 in leukemia.

Abstract Human B7 homolog 4 (B7-H4) is a transmembrane protein in the B7 family of molecules that is expressed on tumor cells in the tumor microenvironment (TME). High protein expression on tumors and low expression on healthy tissue makes B7-H4 an attractive molecule for direct targeting with low off-target toxicity. NC762 is a humanized immunoglobulin gamma 1, kappa (IgG1κ) monoclonal antibody specific for human B7-H4 that is being developed for the treatment of cancer and demonstrates an excellent safety profile in IND-enabling studies. The IgG1κ region of NC762 contains three-point mutations (S239D/A330L/I332E a.k.a. DLE) which enhances binding to CD16a (FcγRIIIa) in order to increase antibody-dependent cellular cytotoxicity (ADCC) activity. NC762 binds to human B7-H4 on SKBR3 cells, a breast cancer cell line that endogenously expresses B7-H4 and induces in vitro ADCC activity. Despite the high homology, NC762 does not cross-react with rodent B7-H4, therefore, in vivo testing was performed in a humanized xenograft murine model in NSG mice, where NC762 restricted tumor growth in a dose-dependent manner. Depleting human natural killer (NK) cells prior to tumor inoculation reduced anti-B7-H4 parent antibody activity, supporting a role for NK-cell mediated ADCC in tumor growth retardation. However, in the absence of human PBMCs, NC762 also achieved significant anti-tumor activity. Furthermore, a mutant of NC762 with a low FcγR binding IgG1 domain showed similar activity as the DLE IgG1, demonstrating an ADCC-independent mechanism of tumor growth restriction mediated by B7-H4 mAb treatment. Together, these results indicate that even with the ADCC-enhanced IgG1κ region, NC762 is able to mediate ADCC-independent anti-tumor activity for the eradication of B7-H4 expressing tumors; and is expected to provide benefit in multiple oncology indications. Citation Format: Kristina Archer, Justine Ceradoy, Tiffany Coupet, Jason Bosiacki, Chang Song, Ido Weiss, Sue Niu, Tete Obot, Jinglin Xu, Tom O'Neill, Alison Ferrell, Hasan Abukharma, Zac Cusumano, Dallas Flies, Sol Langermann, Linda N. Liu. Development and functional characterization of NC762, a novel therapeutic antibody targeting B7-H4, for the treatment of malignancies [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2021; 2021 Apr 10-15 and May 17-21. Philadelphia (PA): AACR; Cancer Res 2021;81(13_Suppl):Abstract nr 3193.

Despite advances that have greatly improved the overall survival of pediatric B cell acute lymphoblastic leukemia (B-ALL), it remains one of the leading causes of cancer-related death in children. Immunotherapy has shown efficacy in treatment of refractory disease, highlighting the need for greater understanding of the immune evasion mechanisms underlying this disease so that additional immune modulating therapeutic strategies can be developed. Siglec-15 (Sig15) was recently reported to have immune modulatory functions in the context of solid tumors. We have found that SIGLEC15 is overexpressed at the RNA level in primary B cell acute lymphoblastic leukemia (B-ALL), acute myelogenous leukemia (AML), and diffuse large B cell lymphoma as compared to healthy donor controls. As compared to healthy donor PBMCs, we have confirmed higher expression of Sig15 at the RNA and protein levels through RT-qPCR, immunoblotting, and flow cytometry across a panel of human B-ALL, AML, DLBCL, and T cell acute lymphoblastic leukemia (T-ALL) cell lines. Knockout of Sig15 expression in a BCR-ABL1+ murine model of B-ALL engrafted in immunocompetent and Rag1-/- immunodeficient recipients resulted in leukemia clearance in immunocompetent, but not immunodeficient, recipients and 100% survival (Figure 1). These data suggest a prominent role for Sig15 in the suppression of adaptive immune response to B-ALL as well as other hematological malignancies. Additional studies suggest that SIGLEC15 expression is positively regulated by NFκB signaling, which is known to be constitutively activated in certain B-ALL subsets. Importantly, we have observed release of a soluble form of Sig15 (sSig15) from B-ALL cells, which is regulated by PKC and calcineurin-mediated signaling. To discover translational application, we measured sSig15 in the plasma of both healthy and pediatric leukemia patients and found significantly higher levels of sSig15 as compared to healthy individuals (Figure 2; LLD = 5 pg/ml; **P<0.01). Together, these results suggest Siglec-15 is a novel and potent immunosuppressive molecule active in leukemia that may be targeted therapeutically to activate lymphocytes against leukemia cells. Disclosures Abukharma: NextCure, Inc.: Current Employment. Liu:NextCure, Inc.: Current Employment.

BackgroundSiglec-15 (S15) is a member of the Siglec family of immunoglobulin superfamily proteins involved in immune regulation. NC318 is a first-in-class humanized IgG1κ monoclonal antibody that blocks S15-mediated immune suppression.MethodsThe Phase 1 dose-escalation study was a classical 3+3 design in 15 tumor types (n=49). Phase 2 (n=47) was conducted at 400 mg q2w in 4 tumor types. Inclusion criteria included subjects with advanced/metastatic solid tumors refractory or resistant to currently available therapies with a TPS PD-L1 score <50%. The median number of previous therapies was ≥3, including checkpoint inhibitors (figure 1).ResultsNC318 was well tolerated with no novel immunologic or safety signals observed. Disease control rate amongst evaluable population (n=83) is 38% {1 CR, 3 PR and 28 SD (stable disease)}. Median duration of disease control is 24 weeks (16–48 weeks) amongst 20 subjects achieving a minimal 16-week duration of stable disease. Two NSCLC subjects (1CR and 1PR) are still on therapy over 2 years. We observed an increase in a soluble form of Siglec-15 (sS15) in all patients receiving NC318 treatment that was dose-dependent. sS15 serves as a pharmacodynamic marker for NC318 activity. PK/PD modeling of NC318 from this Phase1/2 study using sS15 as a PD marker suggested increasing the dose of NC318 to 800 mg q1w to enhance overall exposure of NC318. Development of an S15 specific IHC assay allowed us to do post-hoc analysis by immuno-histochemistry (IHC) from screening biopsies amongst subjects who showed disease control (CR, PR and SD) compared to subjects with progressive disease. S15 expression on tumor cell membrane was a predictor for stable disease, longer duration on therapy when compared to progressive disease {H score ≥ 1 (p=0.046), including NSCLC subjects}, as well as for progression-free survival (PFS) (figures 2 and 3). There was no correlation with the outcome whether PD-L1 was positive or negative. Together, development of a predictive indicator of S15 staining coupled with the NC318 PK/PD data, resulted in a protocol amendment to prospectively enroll subjects with Siglec-15+ adenocarcinoma lung, squamous H&N, and breast cancers at 800 mg q1w. Soluble S15, immunophenotyping, cytokine and chemokine levels and neutrophil-lymphocyte ratio will be presented at the meeting.Abstract 490 Figure 1NC318: study schemaAbstract 490 Figure 2Tumor membrane S15 H score≥1 and progression-free survival. A) All Individuals with available H-Scores>=1 were stratified into two groups (Progressive disease (PD) and Stable disease (SD)) based on the RECIST criteria and their plasma membrane H-scores were compared using Wilcoxon test. Significant differences among H-scores were observed between the groups with a p-value of 0.046; B) Survival analysis was performed by stratifying individuals with H-Scores>=1 into two groups (PD or SD). Statistical analysis was performed by Log-rank (Mantel-Cox) and Hazard Ratio (Mantel-Haenszel) test.Abstract 490 Figure 3Tumor membrane S15 H score and PFS (SD vs. PD). A) All Individuals with available H-Scores were stratified into two groups (Progressive disease (PD) and Stable disease (SD)) based on the RECIST criteria and their plasma membrane H-scores were compared using Wilcoxon test. Significant differences among H-scores were observed between the groups with a p-value of 0.046. All individuals with H-score>=1 are above the dashed red horizontal line; B) Survival analysis was performed by stratifying as CR or PR, SD at week 16, SD at week 8 and PD at week 16, SD at week 8 and did not reach week 16, and PD by RECIST criteria. P value is generated by Log-rank (Mantel-Cox) test between groups of SD (all three subsets combined) vs. PD. Analysis indicates differences in median survival rates with better survival attributed to individuals with response in the above specified orderConclusionsNC318 shows promising early evidence of disease control in subjects with Siglec-15 positive advanced or metastatic solid tumors in phase 1 & 2 studies, prompting evaluation of S15 expression as a predictive biomarker in the prospective study at 800mg q1w dosing.Trial RegistrationNCT03665285Ethics ApprovalThis study has been approved by the IRB of all the participating institutions, and all participants have given informed consent before taking part in the study.ConsentWritten informed consent was obtained from the patient for publication of this abstract and any accompanying images. A copy of the written consent is available for review by the Editor of this journal.

BackgroundCollagen and C1q in the extracellular matrix (ECM), are the predominant ligands for LAIR-1, an inhibitory receptor expressed on the cell surface of several immune cell subsets that inhibits immune activation and migration. LAIR-2, a soluble homolog of LAIR-1, competes for binding to collagen and C1q and serves as a natural decoy to promote immune function under normal conditions. Dysregulation of the ECM and increased expression of collagen and C1q within the tumor microenvironment (TME) plays a critical role in promoting tumor progression. NC410 was engineered to overcome this highly immunosuppressive environment by blocking LAIR-1 function, reversing immune suppression, and inducing ECM remodeling to promote immune cell infiltration within the TME.MethodsThis is a first in human, phase 1/2, open-label, single-armed dose-escalation study to determine the safety, tolerability, dose-limiting toxicity (DLT), maximum tolerated dose (MTD), recommended phase 2 dose, preliminary efficacy and to explore pharmacodynamic biomarkers of NC410 (figure 1). Key eligibility criteria include subjects with advanced or metastatic solid tumors with measurable disease based on RECIST v1.1.ResultsAs of 07/22/2021, a total of 16 patients have been enrolled, treated, and completed the DLT period. NC410 (up to 60 mg), was well tolerated, with no safety concerns, infusion-related toxicities, or DLT reported. No anti-drug antibody (ADA) was detected post-NC410 up to 60 mg treatment. As expected, the C1Q level decreased immediately after the NC410 infusion and was replenished after two hours. Evaluation from samples to date available up to cycle 5 suggests that there was no reduction in the baseline C1Q level with subsequent dosing (figure 2). LAIR-2 levels continued to increase in a dose-dependent fashion post-NC410 dosing and marginal increase pre-dose (figure 3). Interestingly, we observed an increase in soluble LAIR-1 over time in a similar pattern to LAIR-2 (figure 4). Furthermore, immunophenotyping of patient whole blood suggests a trend towards an increase in CD4+ and CD8+ T cells including LAIR-1+ T cells in cohort 4, although overall expression levels of LAIR-1 did not appear to increase (ongoing analysis). Cytokines, chemokines, and collagen degradation products (CDP) will be evaluated as potential pharmacodynamic biomarkers as we continue through higher dose cohorts.Abstract 487 Figure 1NC410: study schemaAbstract 487 Figure 2C1Q levels show immediate reduction after NC410 dosing with no accumulated depletionAbstract 487 Figure 3LAIR-2 levels show an increase in dose-dependent fashion after NC410 dosing with marginal increase pre-doseAbstract 487 Figure 4Soluble LAIR-1 levels show a similar pattern to LAIR-2 levelsConclusionsPreliminary evaluation of NC410 in subjects with advanced or metastatic solid tumors appears to be safe and well-tolerated with evidence of immune modulation consistent with predictive PK/PD modeling in a Phase 1/2 open-label study. Further evaluation will be done with increasing doses to confirm these initial findings.Trial RegistrationNCT04408599Ethics ApprovalThis study has been approved by the IRB of all the participating institutions, and all participants have given informed consent before taking part in the study.ConsentWritten informed consent was obtained from the patient for publication of this abstract and any accompanying images. A copy of the written consent is available for review by the Editor of this journal.