David E. Szymkowski

Supplementary Figure S1: XmAb808 Combines With TCEs to Promote IL2 secretion in Naive, Central Memory, Effector Memory, and TEMRA T Cells.

Supplementary Figure S1: XmAb808 Combines With TCEs to Promote IL2 secretion in Naive, Central Memory, Effector Memory, and TEMRA T Cells.

Supplementary Figure 2. XmAb808 Combines With a B7-H3×CD3 TCE to Stimulate IL2 and IFNγ Release From Chronically Stimulated, Exhausted T Cells.

Supplementary Figure 2. XmAb808 Combines With a B7-H3×CD3 TCE to Stimulate IL2 and IFNγ Release From Chronically Stimulated, Exhausted T Cells.

CD28 costimulation drives activation and survival of T cells, leading to IL2-dependent killing of cancer cells. Target cells were cocultured with human T cells (E:T = 1:1) and treated with XmAb808. A, After 24 hours, IL2 (closed symbols) and IFNγ (open symbols) were measured in culture supernatants. B, The percentages of CD4+ or CD8+ T cells that were CD25+ or Bcl-xL+ from cocultures with 22Rv1-αCD3 cells following 24 hours of treatment are plotted. C, Real-time killing of 22Rv1-αCD3 cells in cocultures treated with (closed symbols) or without (open symbols) 1 µg/mL of XmAb808 plus an IL2-neutralizing antibody, an IFNγ-neutralizing antibody, or an isotype control antibody was monitored over time using an xCELLigence Real-Time Cell Analysis System. The arrow indicates the addition of T cells and antibodies at 24 hours after cancer cells were plated. Data are plotted as single points in A and B and as mean ± SEM of three replicates in C.

XmAb808 binds avidly to B7-H3 on target cells to costimulate T cells. A, Binding of XmAb808 or its analogue containing a single B7-H3–binding domain (mono–B7-H3×CD28) to target cells with varying B7-H3 densities (left axis) and binding of XmAb808 to human T cells (right axis) are plotted as mean fluorescence intensity (MFI). HEK293 cells express approximately 200,000 B7-H3 antigens; A431, 196,000; 22Rv1, 86,000; LOX-IMVI cells do not express B7-H3. Because mono–B7-H3×CD28 has a lower molecular weight than XmAb808 (145 vs. 192 kDa), the mono–B7-H3×CD28 concentrations in this figure were corrected by a factor of 1.3-fold to adjust for molarity. B, Human PBMCs from 20 different donors were incubated on plates coated with air-dried (immobilized) XmAb808, anti-CD28 (TGN1412), or anti-CD3 (OKT3) antibodies. IFNγ and IL2 were measured after 24 hours. Statistical significance is denoted by **, P < 0.01; ns, not significant. C, IL2 release from PBMCs cocultured with HEK293-αCD3 or HEK293-αCD3-B7-H3 KO cells at an E:T ratio of 25:1 was measured after 24 hours. Data are represented as means ± SEM of cocultures with PBMCs from eight different human donors.

Supplementary Figure 2. XmAb808 Combines With a B7-H3×CD3 TCE to Stimulate IL2 and IFNγ Release From Chronically Stimulated, Exhausted T Cells.

T-cell activation is a multistep process requiring T-cell receptor engagement by peptide–MHC complexes (Signal 1) coupled with CD28-mediated costimulation (Signal 2). Tumors typically lack expression of CD28 ligands, so tumor-specific Signal 1 (e.g., neoepitope presentation) without costimulation may be ineffective or even induce T-cell anergy. We designed the bispecific antibody XmAb808 to co-engage the tumor-associated antigen B7-H3 with CD28 to promote T-cell costimulation within the tumor microenvironment. XmAb808 costimulation was measured by its ability to activate and expand T cells and enhance T cell–mediated cancer cell killing in cocultures of human peripheral blood mononuclear cells and cancer cells and in mice engrafted with human peripheral blood mononuclear cells and tumor xenografts. XmAb808 avidly bound cancer cells and stimulated IL2 and IFNγ secretion from T cells cocultured with cancer cells engineered to deliver Signal 1 to T cells via a surface-expressed anti-CD3 antibody. XmAb808 enhanced expression of the antiapoptotic factor Bcl-xL and CD25, promoting survival and IL2-dependent expansion of T cells coupled with increased T cell–mediated cytotoxicity in vitro. XmAb808 combined with an EpCAM×CD3 bispecific antibody to enhance target cell killing through IL2-dependent expansion of CD25+ T cells. This combination also suppressed pancreatic tumor xenograft growth in mice. Furthermore, XmAb808 combined with an anti–programmed cell death protein 1 antibody to suppress breast tumor xenograft growth in mice. XmAb808 as monotherapy and in combination with an anti–programmed cell death protein 1 antibody is currently in clinical development in patients with advanced solid tumors. Our results suggest that XmAb808 may also combine with tumor antigen–targeted anti-CD3 (Signal 1) T-cell engagers.

XmAb808 enhances T-cell recall responses to endogenous pMHC ligands. A (top graph), Levels of HLA-A2 on A431-β2M-null cells expressing a fusion protein of β2M and HLA-A*0201 fused to CMV pp65 NLV peptide (A431-NLV, dashed black) or melanoma G209 peptide (A431-G209, solid gray) were compared with the signal from unstained cells (solid black). A (two bottom graphs), PBMCs from two unique HLA-A*0201 CMV–seropositive donors were stained with HLA-A*0201-CMV-pp65 (NLV) tetramers; the percentage of tetramer+ CD8+ T cells is shown. B, A431-NLV or A431-G209 cells were cocultured with CD3+ T cells from donor 1 at an E:T ratio of 25:1 and treated with XmAb808. IL2 and IFNγ were measured in the supernatants 24 hours after treatment. C, Parental MDA-MB-231 cells or MDA-MB-231 cells expressing the CMV pp65 protein (MDA-MB-231-pp65) were cocultured with CD3+ T cells from donor 2 at an E:T ratio of 10:1 and treated with XmAb808. IL2 and IFNγ were measured in the supernatants 24 hours after treatment. Data are represented as the mean ± SEM of three replicates.

Supplementary Figure S1: XmAb808 Combines With TCEs to Promote IL2 secretion in Naive, Central Memory, Effector Memory, and TEMRA T Cells.