Strategies to target long-lived plasma cells for treating hemophilia A inhibitors

Abstract

Long-lived plasma cells (LLPCs) can persistently produce anti-factor VIII (FVIII) antibodies which disrupt therapeutic effect of FVIII in hemophilia A patients with inhibitors. The migration of plasma cells to BM where they become LLPCs is largely controlled by an interaction between the chemokine ligand CXCL12 and its receptor CXCR4. AMD3100 combined with G-CSF inhibit their interactions, thus facilitating the mobilization of CD34(+) cells and blocking the homing of LLPCs. These reagents were combined with anti-CD20 to reduce B-cells and the specific IL-2/IL-2mAb (JES6-1) complexes to induce Treg expansion for targeting anti-FVIII immune responses. Groups of mice primed with FVIII plasmid and protein respectively were treated with the combined regimen for six weeks, and a significant reduction of anti-FVIII inhibitor titers was observed, associated with the dramatic decrease of circulating and bone marrow CXCR4(+) plasma cells. The combination regimens are highly promising in modulating pre-existing anti-FVIII antibodies in FVIII primed subjects.

Keywords: AMD3100; Factor VIII; G-CSF; Hemophilia A; Immune tolerance; Immunomodulation; Inhibitors; Plasma cells.

Figure 1. FVIII activities and inhibitor titers following treatment with combination regimen in FVIII-plasmid primed HemA inhibitor mice

HemA mice were initially primed with FVIII-plasmids for induction of inhibitor titers. (A, B) One group of inhibitor mice (n=4) were treated with the combination regimen IL-2/IL-2mAb complexes + anti-CD20 + G-CSF + AMD3100; (C, D) the inhibitor mice (n= 4) were setup as the non-treated inhibitor only control group. Combined treatment was scheduled at 2-weeks per cycle for three cycles. Peripheral blood was then collected on weeks 1, 3, 6 and 10 for evaluation of FVIII activities (A, C) and inhibitor titers (B, D) at different time points. 2^nd FVIII-plasmid challenge was scheduled on week 8 when the inhibitor titers were decreased to zero in mice after 3-cycles of combination regimen. Data shown is representative of two independent experiments.

Figure 2. Effects of immunomodulation on CD4⁺ T cells, CD4⁺CD25⁺Foxp3⁺ Tregs cells and T cell activation markers in peripheral blood of combination regimen treated FVIII-plasmid primed inhibitor mice on weeks 1, 3, 6 and 10

Lymphocytes were isolated from peripheral blood of IL-2/IL-2mAb complexes + anti-CD20 + G-CSF + AMD3100 treated inhibitor mice (black bar; n=4), inhibitor only mice (gray bar; n=4) and naive mice (white bar; n=4) at different time points. (A) CD4⁺Foxp3⁺/CD25⁺ Tregs in CD4⁺ T cells and (B) CD4⁺Foxp3⁺/Helios⁺ cells in CD4⁺ T cells were stained and analyzed by flow cytometry during treatment. Blood cells were also stained and analyzed for Treg activation markers: (C) CD25, (D) GITR and (E) CTLA4. Data shown are median fluorescence intensity (MFI) values of the three activation markers, and representative of two independent experiments.

Figure 3. Effects of immunomodulation on total B, mature B, transitional B, memory B, plasma, and CXCR4⁺ plasma cells in peripheral blood of each mouse group

Lymphocytes were isolated from the peripheral blood of IL-2/IL-2mAb complexes + anti-CD20 + G-CSF + AMD3100 combination treated inhibitor mice (black bar; n=4), inhibitor only mice (gray bar; n=4) and naive mice (white bar; n=4) at different time points during the treatment period (week 1, 3, 6 and 10). (A) B220⁺ (total) B cell, (B) IgM⁺IgD^hi (mature) B cells, (C) IgM^hiIgD⁺ (transitional) B cells, (D) IgM⁻IgD⁻ (memory) B cell, (E) B220⁻CD138⁺ PCs as well as (F) CXCR4⁺ specific PCs populations were investigated using flow cytometry analysis. Data shown are cell percentages (A-C, and F) and real numbers (D and E), and representative of two independent experiments.

Figure 4. Inhibitor titers and anti-FVIII IgG1 levels in FVIII protein primed HemA mice with pre-existing inhibitors following combination treatment

Two groups of HemA mice were treated separately with different combination regimens 2-weeks per cycle for 3 cycles: (A) IL-2/IL-2mAb complexes + anti-CD20 + AMD3100 + G-CSF (n=7), (B) Anti-CD20 + AMD3100 + G-CSF (n=6), and (C) Inhibitor mice only (n=3) as the control group. Peripheral blood was collected at different time points following the combination treatment. Anti-FVIII antibody titers were assessed by Bethesda assay over time. Each symbol represents data obtained from an individual mouse. (D) Anti-FVIII IgG1 levels were examined in each treated mouse group by ELISA. Data shown is representative of two independent experiments.

Figure 5. Effects of immunomodulation of CD4⁺ T cells, CD4⁺CD25⁺Foxp3⁺ s and T cell activation markers in peripheral blood of treated HemA inhibitor mice on weeks 2, 4, 6 and 10

Lymphocytes were isolated from the peripheral blood of mice treated with combination regimens IL-2/IL-2mAb complexes + anti-CD20 + AMD3100 + G-CSF (n=7, black bar) and Anti-CD20 + AMD3100 + G-CSF (n=6, gray bar), and inhibitor only mice (n=3, white bar). Peripheral blood were collected at different time points following the combination treatment.(A) CD4⁺Foxp3⁺/CD25⁺ cells (%) in CD4⁺ T cells and(B) CD4⁺Foxp3⁺/Helios⁺ cells (%) in CD4⁺ T cells were stained and analyzed by flow cytometry during treatment. Blood cells were also stained and analyzed for Treg markers: (C) CD25, (D) GITR and (E) CTLA4. Data shown are MFI values of the three activation markers, and representative of two independent experiments.

Figure 6. Effects of immunomodulation on total B, mature B, transitional B, memory B, plasma B and CXCR4⁺ plasma B cells in peripheral blood of treated HemA inhibitor mice

Lymphocytes were isolated from the peripheral blood of naive (slant bar; n=3), inhibitor only (white bar; n=3), anti-CD20 + G-CSF + AMD3100 (gray bar; n=6) and IL-2/IL-2mAb complexes + anti-CD20 + G-CSF + AMD3100 (black bar, n=7) treated mice. (A) B220⁺ (Total) B cells, (B) IgM⁺IgD^hi (mature) B cells, (C) IgM^hiIgD⁺ (transitional) B cell, (D)IgM⁻IgD⁻ (memory) B cells, (E) B220⁻CD138⁺ PCs and (F) CXCR4⁺ PCs were stained and analyzed by flow cytometry during treatment period (week 2, 4, 6 and 10). Data shown are cell percentages (A-C, and F) and real numbers (D and E), and representative of two independent experiments.

Figure 7. Depletion of FVIII-specific antibody-secreting cells (ASCs) and memory B cells in the inhibitor mice treated with two different combination regimens

FVIII protein-primed inhibitor mice were treated with IL-2/IL-2mAb complexes + G-CSF + AMD3100 + anti-CD20 (black bar, n=2), Anti-CD20 + G-CSF + AMD3100 (gray bar, n=2), inhibitor only control (white bar, n=2), and naïve mice control (slant bar, n=2). Cells were isolated by MACS from spleens (A, C) and BMs (B, D) at 4hrs, 4weeks, 6weeks and 21 weeks following the first combination treatment. 1×10⁶ cells were used to detect FVIII-specific ASC (A, B) and FVIII-specific memory B cells (C, D) by ELISPOT assays. The two mouse groups treated with combination regimens showed dramatically depletion of their ASCs at 4 and 6 weeks both in spleen and BM compare to the inhibitor only control. FVIII-specific memory B cells were significantly depleted on weeks 6 following combination treatment. Data shown are the average spot numbers of ELISPOT results.