Doses and formulations of Anti-il-15 antibody for the treatment of immune diseases

Abstract

Described monoclonal antibody against IL-15, stable formulations, and specific doses thereof. The antibody is suitable for treating immune disorders in humans, including celiac disease, vitiligo, alopecia, and atopic dermatitis. The antibody is further characterized by extended suppression of IL-15 and NK cell numbers in serum, suppression of specific cell types involved in disease, and a lack of serious adverse effects.

Description

DOSES AND FORMULATIONS OF ANTI-IL-15 ANTIBODY FOR THE TREATMENT OF IMMUNE DISEASESREFERENCE TO SEQUENCE LISTING SUBMITTED ELECTRONICALLY

[0001] The content of the electronically submitted sequence listing (Name 2873_381PC03_SequenceListing_ST26.xml; Size: 17,917 bytes; and Date of Creation: April 16, 2025) filed with the application is incorporated herein by reference in its entirety.PRIORITY CLAIM

[0002] This application claims the benefit of priority to U.S. Provisional Application No. 63 / 639,312, filed April 26, 2024; U.S. Provisional Application No. 63 / 690,643, filed September 4, 2024; and U.S. Provisional Application No. 63 / 781,856, filed April 1, 2025, each of which is incorporated herein by reference in its entirety.BACKGROUND

[0003] IL- 15 has been linked to a number of immune diseases, but it often remains to demonstrate that blocking IL- 15 will treat such a disease. Even with such a link demonstrated, it remains to identify a suitable drug, the appropriate dose, and a stable drug formulation that is capable of inhibiting IL-15, treating IL-15-related diseases, and doing so without serious side effects.BRIEF SUMMARY

[0004] Provided herein is an aqueous pharmaceutical formulation, comprising: (a) about 20 mg / mL to about 150 mg / mL of TEV-53408 antibody, and pharmaceutically acceptable carrier The aqueous pharmaceutical formulation can also contain (b) Histidine; (c) Sucrose; (d) Ethylenediaminetetraacetic acid (EDTA); and (e) Polysorbate-80.

[0005] TEV-53408 antibody specifically binds to human IL-15 and comprises: a heavy chain variable region complementarity determining region (CDR)l comprising the amino acid sequence of SEQ ID NO: 1, a heavy chain variable region CDR2 comprising the amino acidsequence of SEQ ID NO: 2, a heavy chain variable region CDR3 comprising the amino acid sequence of SEQ ID NO: 3, a light chain variable region CDR1 comprising the amino acid sequence of SEQ ID NO: 4, a light chain variable region CDR2 comprising the amino acid sequence of SEQ ID NO: 5, and a light chain variable region CDR3 comprising the amino acid sequence of SEQ ID NO: 6. In some aspects, TEV-53408 antibody comprises a heavy chain variable region comprising the amino acid sequence of SEQ ID NO: 7 and a light chain variable region comprising the amino acid sequence of SEQ ID NO: 8. In some aspects, TEV-53408 antibody comprises a heavy chain comprising the amino acid sequence of SEQ ID NO: 9 and a light chain comprising the amino acid sequence of SEQ ID NO: 10.

[0006] In some aspects, TEV-53408 antibody comprises a human IgG4 heavy chain constant region which has one or more amino acid substitutions. In some aspects, the human IgG4 heavy chain constant region comprises an S228P substitution (by EU numbering), a triple substitution of M252Y / S254T / T256E (by EU numbering), or terminal lysine deletion (K447A) (by EU numbering). In some aspects, the human IgG4 heavy chain constant region comprises an S228P substitution (by EU numbering), a triple substitution of M252Y / S254T / T256E (by EU numbering), and terminal lysine deletion (K447A) (by EU numbering. In some aspects, the antibody or antigenbinding fragment further comprises a human lambda light chain constant region.

[0007] The dose concentration can be selected based on a volume that can fit into a prefilled syringe or autoinjector (typically 3.0ml or less) but usually not less than 0.5ml because small volumes can pose challenges to filling and administration, especially with autoinjectors. When antibody is dosed at 10-50mg, the antibody can be in a formulation with a concentration of about 20 mg / ml to about 50 mg / ml, preferably about 20 mg / ml. When dosed at 50mg and above, the antibody can be in a formulation of about lOOmg / ml. Doses at 150mg and above can be in a formulation of about 150 mg / ml. Higher concentration formulations, such as at 150 mg / ml, can also contain a viscosity modifier, such as arginine.

[0008] In some aspects, the pharmaceutical formulation comprises about 20 to about 100 mg / mL of TEV-53408 antibody, preferably about 20 mg / ml, about 50 mg / ml, or about 100 mg / ml. In some aspects, the pharmaceutical formulation comprises about 150 mg / mL of TEV-53408 antibody. Such a formulation can also contain arginine.

[0009] In some aspects, the pharmaceutical formulation comprises (a) TEV-53408 antibody; (b) about 10 mM to about 50 mM Histidine; (c) about 150 mM to about 350 mM Sucrose; (d) about 0.01 mM to about 0.5 mM EDTA; and (e) about 0.1 mg / mL to about 0.9 mg / mL Polysorbate-80. In some aspects, the pharmaceutical formulation comprises about 20 mMHistidine. In some aspects, the pharmaceutical formulation comprises about 150 mM Sucrose. In some aspects, the pharmaceutical formulation comprises about 180 mM Sucrose. In some aspects, the pharmaceutical formulation comprises about 250 mM Sucrose. In some aspects, the pharmaceutical formulation comprises about 0.1 mM EDTA. In some aspects, the pharmaceutical formulation comprises about 0.5 mg / mL Polysorbate-80. In some aspects, the pharmaceutical composition comprises about 100 mg / mL of the antibody, about 20 mM Histidine, about 250 mM Sucrose, about 0.1 mM of EDTA, and about 0.5 mg / mL Polysorbate-80.

[0010] In some aspects, the pharmaceutical composition further comprises about 75 mM to about 180 mM Arginine-Hydrochloride (Arg-HCl).

[0011] In some aspects, the pharmaceutical composition has a pH of 5.2 ± 0.5.

[0012] In some aspects, the pharmaceutical composition has an osmolality of from about200 mOsm / kg to about 350 mOsm / kg at room temperature.

[0013] In some aspects, the pharmaceutical composition has no significant difference in stability after storage at 2-8° C for up to 24 months, 25°C for up to 18 months, or 40°C for up to 6 months. In some aspects, the pharmaceutical composition has:(a), at least about 98% antibody monomer content after storage at 25 °C for up to 3 months;(b). no significant difference in viscosity, hydrodynamic radius, or polydispersity after storage at 2-8 °C for up to 24 months;(c). no significant difference in sub-visible particle content after storage at 2-8°C for up to 24 months, 25°C for up to 18 months, or 40°C for up to 6 months;(d). from about 60% to about 140% relative potency measured by a cell-based potency assay after storage at room temperature for 24 hours, and after storage at 40°C for up to 1 month;(e). no significant difference in thermal stability after storage at 2-8°C for up to 24 months or 25°C for up to 6 months;(f). no significant difference in secondary and / or tertiary protein structure after storage at 2-8°C for up to 24 months or 25°C for up to 18 months; and / or(g). no significant difference in secondary protein structure after storage at 2-8°C for up to 24 months.

[0014] Also provided herein is a container comprising a pharmaceutical composition of TEV-53408 antibody described herein. In some aspects, the container is a glass vial. In some aspects, the container is a glass vial having a fill volume of about 3 mL. In some aspects, the container is a pre-filled syringe. In some aspects, the syringe is a polypropylene or polycarbonatesyringe. In some aspects, the container is a bag. In some aspects, the bag is a polyolefin or polyvinyl chloride bag. In some aspects, the container is an autoinjector.

[0015] Also provided herein are methods of treating immune disease in a subject in need thereof, comprising administering to the subject TEV-53408 antibody at a dose selected from about 10 to about 700 mg.

[0016] In particular embodiments, the dose is about 10 to about 50 mg every 4 weeks. In other embodiments, the dose is about 40 to about 150 mg every 12 weeks. In particular embodiments the dose is selected from a. about 10 mg every 4 weeks, b. about 20 mg every 4 weeks c. about 30 mg every 4 weeks d. about 50 mg every 4 weeks; e. about 50 mg every 12 weeks, and f. about 150 mg every 12 weeks.

[0017] Such doses are suitable for the treatment immune disorders / disease, particularly autoimmune diseases mediated by tissue resident T-cells. In some aspects, the disorders or diseases mediated by tissue resident T-cells include celiac disease, vitiligo, alopecia, atopic dermatitis, eosinophilic esophagitis, Sjogrens syndrome, graft-versus host disease, type I diabetes, myositis, and rheumatoid arthritis. TEV-53408 antibody can be provided in a formulation described herein.

[0018] In some aspects the treatment of immune disease by TEV-53408 antibody is characterized by (a) a reduction in the number of activated CD8 T resident memory cells and effector cells at a site of disease, (b) reduction in circulating CD8 T memory cells (c) reduction in inflammatory cytokines and / or (d) reduction in pathology. In some aspects, treatment does not significantly reduce the number of NK cells resident at the site of disease.

[0019] In some aspects, the treatment of immune disease by TEV-53408 antibody does not cause a decline in circulating NK cell numbers below l%of total lymphocytes. In some aspects, the treatment does not significantly affect the CD56brightand %CD56dimratio in circulating NK cells. In some aspects, the treatment does not significantly affect the NK cell function, as assessed by CD 107a expression or perforin expression in circulating NK cells. In some aspects, the treatment does not cause serious adverse effects.

[0020] In other aspects, provided herein a formulation of TEV-53408 antibody that blocks IL- 15, reduces the number of activated CD8 T resident memory cells at the site of disease, reducesthe number of effector memory T cells at the site of disease, reduces Th2 inflammatory cytokines at the site of disease; and / or reduces central CD8 T memory cells. In other aspects, provided herein is a formulation of TEV-53408 antibody that reduces the number of circulating NK cells, does not block cell activation in remaining NK cells, and does not affect NK progenitor cells.

[0021] In some aspects, the method comprises administration of TEV-53408 antibody for the treatment of celiac disease, wherein the treatment of celiac disease comprises: (a) repairing the mucosa of a small intestine; (b) increasing the mean villous height vs. crypt depths (V / C); and / or (c) decreasing symptoms of one or more of muscle pain, body pain, joint pain, fatigue, bloating, gas, nausea, cramps, constipation, diarrhea, skin rash, headache, migraine headache, depression, anxiety, brain fog, and irritability; as assessed by patient reported outcomes.

[0022] In some aspects, the method comprises administration of TEV-53408 antibody for the treatment of vitiligo, wherein the treatment of vitiligo comprises (a) reduction in the rate and / or area of skin depigmentation, (b) increase in the area of skin pigmentation (c) reduction in the number and / or activity of melanocyte-targeting T-cells in the skin and / or (d) reduction in the number and / or activity of melanocyte-targeting T-cells in other organs. Skin pigmentation can be measured by F-VASI and T-VASI; these should alter with skin repigmentation. In the treatment of vitiligo, the method can further comprise treatment with ultraviolet light.

[0023] In some aspects, the method comprises administration of TEV-53408 antibody for the treatment of alopecia areata, wherein the treatment of alopecia areata comprises (a) reduction in the rate of hair loss, (b) increase in hair number, growth, and / or density (c) reduction in the number and / or activity of hair follicle-targeting T-cells in the skin and / or (d) reduction in the number and / or activity of melanocyte-targeting T-cells in other organs.

[0024] In some aspects, the method comprises administration of TEV-53408 antibody for the treatment of atopic dermatitis, wherein the treatment of atopic dermatitis comprises (a) reduction of erythema (b) reduction in lichenification (c) reduction in skin thickening (d) reduction in hyperproliferation and expansion of epidermal cells (as evident by e.g. Ki67 and Krtl6 staining) (e) increased expression of filaggrin (f) reduction in transepithelial water loss (TEWL) (g) reduction the levels of CD4+ and CD8+ T cells in the skin and / or (h) suppression of IL-4, IL-13, IL-31, TSLP and / or TARC.BRIEF DESCRIPTION OF THE FIGURES

[0025] Figure 1 shows Zeta potential (mV) and average hydrodynamic diameter (nm) measurements of TEV-53408 drug substance during a titration from pH 6.17 to pH 8.7.

[0026] Figures 2A-2B show the thermal stability of 11 TEV-53408 formulations (F01-F11) by hydrodynamic radius (Rh) (FIG. 2A) and polydispersity (%Pd) (FIG. 2B).

[0027] Figure 3 shows the prediction profile analysis for all parameters as listed in Table 11 with the maximized desirability factor values for arginine, NaCl, sucrose, histidine, and measured pH shown.

[0028] Figures 4A-4B show the thermal stability of two TEV-53408 formulations (D01 and D02) by hydrodynamic radius (Rh) (FIG. 4A) and polydispersity (%Pd) (FIG. 4B).

[0029] Figures 5A-5B show deconvolution results of an TEV-53408 thermogram with non- 2-state model with either 3 transitions (FIG. 5A) or 4 transitions (FIG. 5B).

[0030] Figure 6 shows DSC analysis of the first two domains of TEV-53408.

[0031] Figure 7 shows the full thermal stability profile for TEV-53408 in D01.

[0032] Figures 8A-8B show near UV CD spectra overlaid with 95% confidence intervals for 3 month and 6 month storage conditions (FIG. 8 A) or 18 month and 24 month storage conditions (FIG. 8B).

[0033] Figure 9 shows an overlay of DSC thermograms grouped by storage temperature before and after processing of the thermograms.

[0034] Figure 10 shows stability sample thermograms at 3, 6, 12, 18 and 24 months at the different storage conditions overlaid with the 95% confidence interval.

[0035] Figure 11 shows the measured viscosity versus protein concentration for TEV- 53408 formulations including Arginine (130 or 200 mM), Histidine 120mM, MgCl 50mM, Proline 1 lOmM, or Leucine 116mM with drug product limits set of either 30 cP or <70 cP.

[0036] Figures 12A-12B show the thermal stability of high concentration TEV-53408 formulations by hydrodynamic radius (Rh) (FIG. 12A) or polydispersity (Pd%) (FIG. 12B).

[0037] Figure 13 shows the mean (+ standard deviation) serum TEV-53408 concentrations (pg / ml) versus time on a linear scale by single dose cohorts up to day 190 (safety analysis set). N=number; sc=subcutaneous; SD=single dose.

[0038] Figure 14 shows the mean (+ standard deviation) serum TEV-53408 concentrations (pg / ml) versus time on a semi-log scale by single dose cohorts up to day 666 (safety analysis set). N=number; sc=subcutaneous; SD=single dose.

[0039] Figure 15 shows the mean (+ standard deviation) serum TEV-53408 concentrations (pg / ml) versus time on a semi-log scale by multiple dose cohorts up to day 246 (safety analysis set). MD=multiple dose; N=number; sc=subcutaneous; q4w=every 4 weeks.

[0040] Figure 16 shows the mean (+ standard deviation) serum TEV-53408 concentrations (pg / ml) versus time on a linear scale by multiple dose cohorts up to day 246 (safety analysis set). MD=multiple dose; N=number; sc=subcutaneous; q4w=every 4 weeks.

[0041] Figure 17 shows the mean (+ standard deviation) free serum IL- 15 concentration (ng / L) in all single dose cohorts versus time on a linear scale by treatment up to day 190 (pharmacodynamic analysis set). Free IL-15 is the IL-15 / IL-15Ra complex. SD=single dose

[0042] Figure 18 shows the mean (+ standard deviation) free serum IL- 15 concentration (ng / L) in all single dose cohorts versus time on a linear scale by treatment up to day 666 (pharmacodynamic analysis set). Free IL-15 is the IL-15 / IL-15Ra complex. SD=single dose

[0043] Figures 19A-19C show the mean (+ standard deviation) free serum IL- 15 concentration (ng / L) in multiple dose (MD) cohorts versus time on a linear scale by treatment up to day 360 (FIGs. 19A and 19B) (pharmacodynamic analysis set). Free IL-15 is the IL-15 / IL-15Ra complex. FIG. 19C shows a comparison of peak levels of free IL- 15 in those administered a single dose (SD) of 5 or 10 mg against multiple doses (MD) of 7.5 mg.

[0044] Figure 20 shows the mean (+ standard deviation) total IL- 15 concentration (ng / L) in all single dose cohorts versus time on a linear scale by treatment up to day 666 (pharmacodynamic analysis set). Total IL- 15 is the sum of Free IL- 15 + TEV-53408-bound Free IL-15.

[0045] Figures 21A-21B show the mean (+ standard deviation) total IL-15 concentration (ng / L) in all multiple dose cohorts versus time on a linear scale by treatment up to day 246 (FIG. 21A) and day 360 (FIG. 21B) (pharmacodynamic analysis set). Total IL-15 is the sum of Free IL- 15 + TEV-53408-bound Free IL-15.

[0046] Figure 22 shows the mean absolute NK cell counts in all single dose cohorts versus time on a linear scale by treatment up to day 190 (safety analysis set). C1=SD cohort 1, C2=SD cohort 2; N=number; sc=subcutaneous; SD=single dose.

[0047] Figure 23 shows the mean absolute NK cell counts in 50 and 150 mg single dose cohorts versus time on a linear scale by treatment up to day 666 (safety analysis set). C1=SD cohort 1, C2=SD cohort 2; N=number; sc=subcutaneous; SD=single dose.

[0048] Figures 24A-24B show the mean absolute NK cell counts in all multiple dose cohorts versus time on a linear scale by treatment up to day 360 (FIG. 24B) (safety analysis set).M1=MD cohort 1, M2=MD cohort 2; MD=multiple dose; N=number; q4w=every 4 weeks; sc=subcutaneous; ULN=upper limit of normal.

[0049] Figure 25 shows a simulated TEV-53408 concentration profile for a multiple dose administration cohort administered 150 mg every 12 weeks. Dashed line represents median concentration profile of 150 mg single dose; solid line represents the median concentration profile of 150 mg ql2w. The shaded area represent the 90% confident interval of simulations.

[0050] Figure 26 shows the simulated median NK cell counts for 150 mg ql2w on 2 administrations vs simulated median NK count for single administration of 150 mg TEV-53408. Dashed line represents median NK cell count profile at the 150 mg single dose; solid line represents the median NK cell count profile after 150 mg ql2w over two doses: Shaded areas represent 90% confident interval. Dashed horizontal line represents LLOQ = 24 10A6 / L.

[0051] Figure 27 shows the %CD56brightof circulating NK cells in 15 subj ects administered placebo, over 204 days. While the average was 9.7%, the standard deviation was high (std. dev 8.3), reflecting both inter-individual variability and intra-individual variability over time.

[0052] Figures 28A-28F shows the distribution of CD107a expression in NK cells before and after in-vitro stimulation by phorbol 12-myristate 13-acetate (PMA) / ionomycin, from individuals treated with a single TEV-53408 dose of Img (FIG. 28A), 5 mg (FIG. 28B), or lOmg (FIG. 28C); with multiple TEV-53408 doses of 2.5mg (FIG. 28D), 7.5mg (FIG. 28E), or placebo (FIG. 28F). The degranulation ability (shown by CD 107a expression) of NK cells was not significantly affected by different doses of TEV-53408, multiple doses of TEV-53408, or over time.

[0053] Figures 29A-29F show the effects of removing IL- 15 on decreasing Stage 4 and 5 NK cells in a bone marrow on a chip model. FIG. 29A shows the effects of IL-15 removal from the cell culture medium between day 28 and day 35 on Stage-specific (per Abel et al (2018). Front. Immunol 9: 1869) NK-cell counts in circulation and in the ceramic scaffold relative to control circuits with continuous IL- 15 supplementation. Mean values ± s.e.m of three donors (N=3) with two chips (n=2) are shown. Comparison of early (stage 1-3) and late (stage 4-6) NK cell counts by Welch's t-test with Holm-Sidak correction, n(circulation)=6, n(ceramic)=5, *p < 0.05 and ns (p > 0.05). FIG. 29B shows sampled cell counts of the lymphoid / NK-cell lineage from the circulation over a timeframe of 49 days. Mean values ± SD of individual cell populations are shown as stacked bar graphs from one CD34+ donor with three chips (N=l, n=3). FIG. 29C shows exemplary flow cytometry plots of CD56 and CD16 expression in CD161+ cells in the circulation and in the ceramic scaffold, used for gating of Stage 3, Stage 4 and Stage 5 NK-cell populationson day 49 of the assay. FIG. 29D shows the fraction of CD 16+ cells (Stage 5 NK-cells) of all CD161+ CD56+ NK-cells at day 35 and day 49 sampled from the circulation or harvested from the ceramic scaffold. Individual chips and mean ± s.e.m are shown (N=2, n=3). Repeated measures mixed effects model REML and Sidak multiple comparisons test, n(day 35)=6, n(day 49)=3, ****p < 0.0001, ***p < 0.001, **p < 0.01 *p < 0.05 and ns (p > 0.05). FIG. 29E shows an exemplary flow cytometry plot of CD 107a expression on Stage 4 NK-cells with and without PMA / ionomycin stimulation. FIG. 29F shows the mean fluorescence intensity of CD107a-BV421 on Stage 2B, Stage 4 and Stage 5 NK-cells at day 35 with and without PMA / ionomycin stimulation. Chip-specific intensity values and geometric mean ± SD are shown from one experiment with two CD34+ donor with three chips (N=2, n=3). Log transformed MFI values were compared by oneway ANOVA and Tukey multiple comparisons test, n=6, ****p < 0.0001, ***p < 0.001, **p < 0.01 *p < 0.05 and ns (p > 0.05).

[0054] Figures 30A-30G show the effects of TEV-53408 antibody on Stage-specific NK- cell counts. FIG. 30A shows the treatment effect at day 35 after 7 or 14 days of treatment with 250 pg / mL TEV-53408 in the circulating cell pool and in the ceramic scaffold population. Mean values ± s.e.m. of three donors (N=3) with two chips (n=2) are shown. Raw cell counts were compared by a repeated measures mixed effects model REML with the Geisser-Greenhouse correction and Dunnett’s multiple comparisons test and Dunnet’s multiple comparisons test, n=6 chips, **p < 0.01 *p < 0.05 and ns (p > 0.05). FIG. 30B shows the relative cell counts of Stage 4 NK-cells in circulation on day 31 and day 35 over a concentration range of TEV-53408 from 0.25 to 250 pg / ml normalized to the Stage 4 NK-cells counts sampled from the same circuit at day 28. Individual circuits and mean values ± s.e.m. of two donors with three chips are shown (only one donor for day 31 analysis). Concentration response curves were calculated as a three parameters curve with a least squares regression. IC50 values were compared with an extra sum-of-squares F test. FIG. 30C shows the proliferation rate of Stage 4 NK-cells in circulation on day 31 and day 35 measured by EdU staining over a concentration range of TEV-53408 from 0.25 ng / ml to 250 pg / ml. Mean values ± s.e.m of three chips (n=3) from one donor are shown. Statistical comparison of means was performed by a one-way ANOVA plus Dunnett’s post hoc test with the IgG isotype condition as control condition, n=3 chips, ****p < 0.0001, ***p < 0.001, **p < 0.01 *p < 0.05 and ns (p > 0.05). FIG. 30D shows the time schedule of the antibody application, cell sampling and start of a recovery phase after day 35 of the assay. FIG. 30E shows the cell counts of Stage 4 NK-cells in circulation sampled from untreated control circuits and circuits treated from day 28 to day 35 with different concentrations of TEV-53408 and allowed to recover following washout andmedium exchange between day 35 to day 49. Measured cell counts were normalized to the cell counts in the same circuit at day 28. Mean values ± SD of three chips of one donor are shown. FIG. 30F shows the cell counts of Stage 4 NK-cells in the ceramic scaffold sampled from untreated control circuits and circuits treated from day 28 to day 35 with different concentrations of TEV- 53408 and allowed to recover following washout and medium exchange between day 35 to day 49. Mean values ± s.e.m. of three chips (n=3) of one representative donor are shown. Statistical comparison of raw cell counts was performed by a one-way ANOVA plus Dunnett’s post hoc test with the IgG isotype condition as control condition, n=3 chips, ***p < 0.001, **p < 0.01 *p < 0.05 and ns (p > 0.05). FIG. 30G shows the CD107a mean fluorescence intensity of Stage 4 NK cells in circulation on day 35, day 42, day 49 and in the ceramic scaffold versus concentration of TEV- 53408. Mean values ± s.e.m. of three chips (n=3) before and after stimulation with PMA / ionomycin are shown.

[0055] Figures 31A-31B show the concentration over time of total IL-15 (free and TEV- 53408 bound, top line) and free IL-15(bottom line) in hIL-15 KI mice after administration of TEV- 53408 at Img / kg (FIG. 31 A), or 0.15 mg / kg (FIG. 3 IB). Free and total IL-15 show an inverted relationship with a slight time shift.

[0056] Figures 32A-32B show the induction of vitiligo in a mouse model. FIG. 32A shows the schedule of TRP2 immunization, leading to depigmentation in the tail, administration of TEV- 53408, and subsequent analysis. FIG. 32B shows (L-R) images of mouse tails bright field tail skin sections, and immunofluorescence stain for TRP2 protein in mouse tail skin section. DAPI is used as counterstain depicting nuclei). The top images are baseline, and show baseline tail pigmentation, the presence of melanocytes in skin epidermis, and TRP2 staining. The bottom three images show depigmentation in the tail; lack of melanocytes in the skin epidermis, and lack of TRP2 staining.

[0057] Figures 33 A-33C show that TEV-53408 reverses vitiligo as shown by pigmentation and changes to CD8 TRM cells. FIG. 33A shows that compared to naive mice, total pigmentation was lower after immunization for TRP2, and pigmentation loss was significantly attenuated by treatment with TEV-53408 compared to saline (shown is average ± SEM. Statistics: unpaired t- test, parametric, two tails, *P<0.05). FIG. 33B shows that vitiligo induced by TRP2 immunization was accompanied by an increase in TRM cells. TRM counts were significantly reduced (**P<0.01) by antibody compared to saline (shown is average ± SEM). FIG. 33C show the number of IFNv (left) and TNFa (right) secreting cells in mice. Skin tissue was digested, total cells, containingCD8+ T cells, extracted, and stimulated ex vivo with either TRP2 or DMSO as a control. The in- vivo setting is shown in the first line under the plot, distinguishing between naive mice, those immunized with TRP2 and then administered saline, and those immunized with TRP2 and then administered antibody. Upon culture of CD8 T cells in vitro, IFNy and TNFa expression were induced by TRP2 activation but not by DMSO. 5mg / ml TEV-53408 treatment in vivo significantly (p<0.05) decreased the number of ITNy secreting cells in response to TRP2, but not the number of TNFa secreting cells (shown is average ± SEM).

[0058] Figures 34A-34C show the effects of TEV-53408 treatment on CD8 T cells. FIG. 34A shows that 5mg / kg TEV-53408 significantly (p<0.01) reduces the number of memory precursor CD8+ T cells in the blood. FIGs. 34B-34C shows that 5mg / kg and Img / kg, but not 0.2 mg / kg, were sufficient to significantly reduce skin CD8 cells (FIG. 34B) and CD8 TRM (FIG. 34C) in the vitiligo model. Shown is average ± SEM, statistics: one-way ANOVA, *>0.05, **>0.01, ***>0.001.

[0059] Figures 35A-35D Analysis of splenocytes and serum of hIL-15KI mice 7 days after a single dose of TEV-53408, COMP1 or COMP2 at different doses, show TEV-53408 is more potent than other clinical-stage anti-IL-15 antibodies in healthy hIL-15KI mice. COMP1 antibody contains the primary sequence of AMG-714 and COMP2 contains the primary sequence of CALY-002, produced based on published patent sequences. FIG. 35A Measurement of free IL-15 (unbound to antibody) levels in the serum (FIG. 35B) Flow cytometry analysis of NK cells (NK1.1+ cells) in the spleen. (FIG. 35C) Flow cytometry analysis of T cells (CD3+ cells) in the spleen. (FIG. 35D) Flow cytometry analysis of CD8+ T cells (CD3+ CD8+ cells) in the spleen.

[0060] Figure 36A-36G. FIG. 36 A. Experiment layout of a Vitiligo efficacy comparative study. Vitiligo was induced in hIL-15KI mice by 4 weekly immunizations on weeks 0-3. TEV- 53408, COMP1 or COMP2 were administered 3 times at 1 mg / kg, once a week on weeks 5-7. TEV-53408 and COMP1 were administered sc while COMP2 was administered ip. Experiment was terminated 24 hours after week 7 timepoint. n=8-10 / group. FIG. 36B. Depigmentation quantification of the tail skin at endpoint. Pixels of tail images were calculated using a MATLABbased program and a constant threshold. Negative values represent loss of pigment. Unpaired t- test: **<0.01. FIGs. 36C-36D. Flow cytometry analysis of immune cells in tail skin epidermis (FIG. 36C) CD8 TRMs (CD103+) (FIG. 36D) CD8 short-liver effector T cells (CD44+, KLRG1+, CD127-). Unpaired t-test: *p<0.05, **<0.01. (FIG. 36E-36F) Flow cytometry analysis of immune cells in the blood: (FIG. 36E) CD8 TRM progenitors (CD44+, CD62L+, KLRG1-, CD127+), (F)CD8 short-liver effector T cells (CD44+, CD62L-, KLRG1+, CD127-). One-way ANOVA: **p<0.01, ****p<0.0001. FIG. 36G. Measurement of free IL-15 (unbound to antibody) levels in the serum of mice at termination. One-way ANOVA, ****p<0.0001.

[0061] Figures 37A-37B. Flow cytometry analysis of NK cells in tail skin epidermis (FIG. 37A) and in the blood (FIG. 37B). Shown is % of NK1.1+ cells out of live. n=8-10 / group, oneway ANOVA, *p<0.05, ****p<0.0001.

[0062] Figures 38A-38C. TEV-53408 efficacy in an alopecia areata mouse model. FIG. 38A Experiment layout of TEV-53408 efficacy study in alopecia areata. Humanized Alopecia mouse model was established by transplantation of human scalp grafts onto the back of Beige- SCID immunodeficient female mice, and injection of NKG2D+ CD8+ autologous T cells into the graft. TEV-53408 was administered sc twice-a-week at 5 mg / kg, at either a preventative regimen (treatment started with CD8+ T cell injection) or a therapeutic regimen (treatment started after Alopecia was established, at day 52). PBS and anti-KLH IgG4 (at 5 mg / kg, sc) were administered as negative controls. Tofacitinib (JAK inhibitor, 5 mg / kg, 5 days a week, po) was administered as a positive control. Quantification of the number of hair per graft in preventative (FIG. 38B) and therapeutic (FIG. 38C) groups. The value for each mouse is the average of 3 transplanted grafts / mouse. n=12 / group, shown is average ± SEM. One-way ANOVA, *p<0.05, **p<0.01, ****pO .0001. Quantification was performed on day 0 (scalp transplantation day), on day 45 or 52 (Alopecia peak) and on day 122 (termination day).

[0063] Figures 39A-39K. TEV-53408 reduces AD phenotypes in a humanized mouse model. FIG. 39A. Experiment layout of TEV-53408 efficacy study in atopic dermatitis (AD). Humanized AD mouse model was established by transplantation of human skin onto the back of Beige-SCID immunodeficient female mice, and injection of Th2-differentiated autologous T cells into the graft. TEV-53408 was administered sc 3 times / week at 2.5 mg / kg, in a preventative regimen, starting at the day of Th2 cell injection (day 28). PBS and anti-KLH IgG4 (at 2.5 mg / kg, sc) were administered as negative controls. Dupilumab (anti-IL-4R antibody, 25 mg / kg, 3 times / week) was administered as a positive control. FIG. 39B. A representative picture of a skin graft from each group at termination. FIG. 39C. Histological evaluation of skin grafts for AD phenotypes, based on H&E staining (n=8). FIG 39D-39F. Scoring of AD phenotypes in the graft: erythema (FIG. 39D), lichenification (FIG. 39E), and crusting (FIG. 39F). n=14 / group, shown is average ± SEM. Kruskal-Wallis ANOVA, *p<0.05, **p<0.01. FIGs. 39G-39H. Immunohistochemical evaluation of the skin sections showed that TEV-53408 reducedhyperproliferation and expansion of epidermal cells (as evident by Ki67 and Krtl6 staining), thus leading to reduction in epidermal thickness, a hallmark of AD. Immunohistochemistry staining of Ki67 (FIG. 39G) and Keratin 16 (Krtl6, FIG. 39H) in the epidermis. Staining was quantified by image J as percent positive cells and percent coverage, respectively. FIG. 391. Epidermal thickness of skin grafts. Thickness was determined using an ocular micrometer at a minimum of 40-50 points along the epidermis selected to represent points of maximal and minimal thickness. FIG. 39J. Transepidermal water loss (TEWL) measurement. TEWL values are expressed in grams per hour per square meter (g / h / m 2). The TEWL probe was connected to an adaptor and the readings were recorded every second interval for at least 20 seconds, using the Tewameter®TM Nano device and software. FIG. 39K. Immunohistochemistry staining of Filaggrin in the epidermis. Staining was quantified by image J as percent coverage.

[0064] Figures 40A-40G. TEV-53408 reduces immune cells and cytokines related to AD phenotypes in a humanized mouse model. FIGs. 40A-40E. Immunohistochemistry of CD4 (FIG. 40A), CD8 (FIG. 40B), IL-4 (FIG. 40C), IL-13 (FIG. 40D) and IL-31 (FIG. 40E) in the epidermis and dermis. Staining was quantified by counting positive cells in an area of 0.66 mm2. FIG. 40F- 40G. Immunohistochemistry of TSLP (FIG. 40F) and TARC (FIG. 40G) in the epidermis. Staining was quantified by image J as percent coverage.DETAILED DESCRIPTIONDefinitions

[0065] In order that the present disclosure can be more readily understood, certain terms are first defined. As used in this application, except as otherwise expressly provided herein, each of the following terms shall have the meaning set forth below. Additional definitions are set forth throughout the application. In the absence of specific definition, terms are to be given their ordinary meaning in the art.

[0066] The term “antibody” is well known in the art. As used herein, “TEV-53408” or “TEV-53408 antibody” (sometimes abbreviated to Antibody, with a capital) herein means the highly potent, fully human immunoglobulin G (IgG) subclass 4 (IgG4) (lambda) monoclonal antibody (mAb) that binds with high affinity to human, rhesus, and cynomolgus monkey IL-15 or IL-15:IL-15Ra complex in US Patent No. 11,267,883. TEV-53408 comprises complementarity determining regions: respectively H-CDR 1 to 3 of SEQ ID: Nos 1-3, and L-CDR 1-3 of SEQ ID NOs: 4-6. TEV-53408 can also be defined by its heavy and light chain variable regions, whichcomprises the heavy chain of SEQ ID NO: 7 and the light chain of SEQ ID NO: 8. The heavy chain variable region is joined to a human IgG4 heavy chain constant region comprising an S228P substitution (by EU numbering), a triple substitution of M252Y / S254T / T256E (by EU numbering), and terminal lysine deletion (K447A) (by EU numbering). TEV-53408 also comprises a human lambda light chain constant region having SEQ ID NO: 12. TEV-53408 can also be characterized by the heavy chain of SEQ ID NO: 9 and the light chain of SEQ ID NO: 10. TEV-53408 acts by inhibiting the binding of the IL-15 to its specific signaling receptors, IL-2RP and fL-2Ry, thereby preventing the formation of the IL- 15 receptor complex, but does not block binding to IL-15Ra chain (IL-15Ra).

[0067] “Binding affinity” refers to the strength of the sum total of non-covalent interactions between a single binding site of a molecule (e.g., an antibody or antigen-binding fragment thereof) and its binding partner (e.g., an antigen). Unless indicated otherwise, as used herein, “binding affinity” refers to intrinsic binding affinity which reflects a 1 : 1 interaction between members of a binding pair (e.g., antibody or antigen-binding fragment thereof and antigen). The affinity of a molecule X for its partner Y can generally be represented by the dissociation constant (KD). Affinity can be measured and / or expressed in a number of ways known in the art, including, but not limited to, equilibrium dissociation constant (KD), and equilibrium association constant (KA). The KD is calculated from the quotient of koff / kon, whereas KA is calculated from the quotient of koff / kon. Konrefers to the association rate constant of, e.g., an antibody or antigen-binding fragment thereof to an antigen, and koff refers to the dissociation of, e.g., an antibody or antigen-binding fragment thereof from an antigen. The konand koff can be determined by techniques known to one of ordinary skill in the art, such as BIAcore® or KinExA.

[0068] The term “isolated” refers to a form which is not found in nature. As used herein, “substantially pure” refers to material which is at least 50% pure (i.e., free from contaminants), at least 90% pure, at least 95% pure, at least 98% pure, or at least 99% pure.

[0069] It is understood in the art that the cell culture production of a monoclonal antibody will result in “main species” of antibody and a minority of variants with, for example, different glycosylation patterns, glycation, oxidation, and alteration of disulfide bonds.

[0070] As used herein, the term “recombinant” includes the expression from genes made by genetic engineering or otherwise by laboratory manipulation.

[0071] As used herein, the term “pharmaceutical formulation” refers to a preparation which is in such form as to permit the biological activity of the active ingredient to be effective, and which contains no additional components which are unacceptably toxic to a subject. The formulation issterile. Typically, such formulations contain a buffer, an isotonicity modifying agent, and a detergent. In one aspect the pharmaceutical formulation is selected from:(a) 100 mg / mL TEV-53408 in 20 mM histidine, 75 mM of arginine hydrochloride, 150 mM sucrose, 0.5 mg / mL PS80 and 0.2 mM of EDTA, at pH 5.85;(b) 100 mg / mL TEV-53408 in 20 mM histidine with 250 mM sucrose, 0.1 mM EDTA, and 0.5 mg / mL polysorbate 80 at pH 5.2; and(c) 150 mg / mL TEV-53408, 20 Mm Histidine, 180 mM Arginine hydrochloride, 180 mM Sucrose, 0.1 mM EDTA, 0.5 mg / mL PS80, pH 5.2.

[0072] The terms “administer,” “administering,” “administration,” etc refer to methods that can be used to enable delivery of TEV-53408 to the desired site of biological action. Administration techniques are found in greater detail in Goodman and Gilman, The Pharmacological Basis of Therapeutics, current edition, Pergamon; and Remington’s, Pharmaceutical Sciences, current edition, Mack Publishing Co., Easton, Pa. and Matucci, A. et al., Respiratory Research, 19(1): 154 (2018).

[0073] As used herein, the terms “combination” or “administered in combination” means that TEV-53408 is administered with one or more additional therapeutic agents.

[0074] The terms “subject” and “patient” are used interchangeably and refer to a human being, unless otherwise specified.

[0075] The term immunologic or immune disorders / disease involve dysregulation of the immune system. As used herein, it refers to diseases of an overactive response of an autoimmune nature, especially those mediated by tissue resident T-cells. “Autoimmune nature” covers diseases like celiac disease which, while triggered by gluten, have pathologic features closer to an autoimmune disease and can continue on a gluten-free diet. Specific disorders or diseases mediated by tissue resident T-cells include celiac disease, vitiligo, alopecia, atopic dermatitis, eosinophilic esophagitis, Sjogrens syndrome, graft-versus host disease, type I diabetes, and rheumatoid arthritis.

[0076] The term “therapeutically effective amount” refers to the amount of TEV-53408 effective to treat a disease or disorder in a subject. Terms such as “treating,” “treatment,” “to treat,” “alleviating,” and “to alleviate” refers to obtaining beneficial or desired results, including therapeutic measures that improve, cure, slow down, lessen symptoms of, and / or halt progression of a pathologic condition or disorder. Those in need of treatment can include those already diagnosed with or suspected of having the disorder. Specific endpoints are defined herein.

[0077] “Specificity” in the context of antibody-antigen interactions is not necessarily an absolute designation but can constitute a relative term signifying the degree of selectivity of an antibody for an antigen-positive cell compared to an anti gen -negative cell. Specificity of an antibody for an antigen-positive cell is mediated by the variable regions of the antibody, and usually by the complementarity determining regions (CDRs) of the antibody.

[0078] As used herein, the term “IL-15” refers to interleukin (IL)-15. Unless otherwise specified it refers to human IL-15. “IL-15 complex” refers to the complex of IL-15 and the IL-15 receptor alpha (IL-15Ra). IL-15 is most commonly found in complex with IL-15Ra. Therefore, measurements of “IL-15” (such as with antibody binding to IL-15), encompass measurements of IL- 15 / IL-15Ra complex, unless otherwise specified

[0079] As used herein, “room temperature” refers to a temperature of from about 15°C to about 25°C, such as about 20°C or about 25°C.

[0080] As used herein, the singular forms “a,” “an,” and “the” include plural forms unless the context clearly dictates otherwise.

[0081] It is understood that wherever aspects of the present disclosure are described herein with the language “comprising,” otherwise analogous aspects described in terms of “consisting of’ and / or “consisting essentially of’ are also provided. Unless specifically stated or obvious from context, as used herein, the term “or” is understood to be inclusive. The term “and / or” as used in a phrase such as “A and / or B” herein is intended to include both “A and B,” “A or B,” “A,” and “B.”

[0082] Where the use of the term “about” is before a quantitative value, the present disclosure also includes the specific quantitative value itself, unless specifically stated otherwise. As used herein, the term “about” refers to a ±10% variation from the nominal value unless otherwise indicated or inferred, with the proviso that if the nominal value is a percentage (%), the value cannot exceed 100%. Thus, for example, the term “about 99%” refers to “89.1% to 100%.”

[0083] As used herein, the term "approximately," as applied to one or more values of interest, refers to a value that is similar to a stated reference value. In some aspects, the term "approximately," like the term, “about,” refers to a range of values that fall within 10%, 9%, 8%, 7%, 6%, 5%, 4%, 3%, 2%, 1%, or less in either direction (greater than or less than) of the stated reference value unless otherwise stated or otherwise evident from the context (except where such number would exceed 100% of a possible value).

[0084] Compositions or methods provided herein can be combined with one or more of any of the other compositions and methods provided herein.

[0085] Units, prefixes, and symbols are denoted in their Systeme International de Unites (SI) accepted form. Numeric ranges are inclusive of the numbers defining the range.Pharmaceutical Compositions

[0086] The present disclosure provides pharmaceutical compositions comprising TEV- 53408 with a pharmaceutically acceptable carrier. The identification of optimal carriers for TEV- 53408 does not preclude the use of other carriers. “Pharmaceutically acceptable” means approved by a government regulatory agency or listed in the U.S. Pharmacopeia or other generally recognized pharmacopeia. “Carrier” refers to a diluent, adjuvant, excipient, or vehicle with which the compound is administered.

[0087] In some aspects, the pharmaceutical formulations provided herein have a pH of about 4.5 to about 6.5. Specific amounts can be a pH of 5.0 ± 0.5, 5.1 ± 0.5, 5.2 ± 0.5, 5.3 ± 0.5, 5.4 ± 0.5, 5.5 ± 0.5, 5.6 ± 0.5, 5.7 ± 0.5, 5.8 ± 0.5, 5.9 ± 0.5, and 6.0 ± 0.5.

[0088] In some aspects, the pharmaceutical formulations provided herein comprise Histidine. Histidine can be present in a range of about 10 mM to about 50mM, including, about 10 mM, about 12 mM, about about 15 mM, about 20 mM, about 25 mM, about 30 mM, about 35 mM, about 40 mM, about 45 mM, or about 50 mM Histidine.

[0089] In some aspects, the pharmaceutical formulations provided herein comprise Sucrose. Sucrose can be present in a range of about 50 to about 300mM, including about 50 mM, about 75mM, about lOOmM, about 125 mM, about 150 mM, about 200mM, about 250mM, and about 275mM Sucrose.

[0090] In some aspects, the pharmaceutical formulations provided herein comprise ethylenediaminetetraacetic acid (EDTA). EDTA can be present in amounts from about 0.05 mM, about 0.1 mM, about 0.2 mM, about 0.3 mM, about 0.4 mM, or about 0.5 mM.

[0091] In some aspects, the pharmaceutical formulations provided herein comprise polysorbate, such as polysorbate 20 or polysorbate 80. In one aspect, it includes about 0.1 mg / mL, about 0.2 mg / mL, about 0.3 mg / mL, about 0.4 mg / mL, about 0.5 mg / mL, about 0.6 mg / mL, about 0.7 mg / mL, about 0.8 mg / mL, or about 0.9 mg / mL polysorbate-80.

[0092] In some aspects, the pharmaceutical formulations provided herein have an osmolality of from about 250 mOsm / kg to about 600 mOsm / kg at room temperature, including from 275 to 325 mOsm / kg, such as about 280-300 mOsm / kg.

[0093] In some aspects, the formulation comprises TEV-53408 in a range of from 20 to 150 mg / ml. Specific formulations include:(a) 100 mg / mL TEV-53408 in 20 mM histidine, 75 mM of arginine hydrochloride, 150 mM sucrose, 0.5 mg / mL PS80 and 0.2 mM of EDTA, at pH 5.85;(b) 100 mg / mL TEV-53408 in 20 mM histidine with 250 mM sucrose, 0.1 mM EDTA, and 0.5 mg / mL polysorbate 80 at pH 5.2; and(c) 150 mg / mL TEV-53408, 20 Mm Histidine, 180 mM Arginine hydrochloride, 180 mM Sucrose, 0.1 mM EDTA, 0.5 mg / mL PS80, pH 5.2.

[0094] In some aspects, the pharmaceutical composition comprises: (a) about 20 mg / mL to about 150 mg / mL of TEV-53408 that comprises (i) H-CDR1 to 3 of SEQ ID: Nos 1-3, and L- CDR of SEQ ID NOs: 4-6, or (ii) a heavy chain of SEQ ID NO: 7 and a light chain of SEQ ID NO: 8; (b) Histidine; (c) Sucrose; (d) Ethylenediaminetetraacetic acid (EDTA); (e) Polysorbate- 80 (PS80) and (f) optionally arginine.

[0095] In some aspects, the pharmaceutical composition comprises: (a) about 10-100 mg / mL of TEV-53408 that comprises (i) H-CDR1 to 3 of SEQ ID: Nos 1-3, and L-CDR of SEQ ID NOs: 4-6, or (ii) a heavy chain of SEQ ID NO: 7 and a light chain of SEQ ID NO: 8; (b) 20 mM histidine, (c) 250 mM sucrose, (e) 0.5 mg / mL PS80 and (f) 0.1 mM of EDTA, at pH 5.2. In some aspects, the composition is about 20, 50 or lOOmg / ml.

[0096] In some aspects, the pharmaceutical composition comprises: (a) about 10-100 mg / mL of TEV-53408 that comprises (i) H-CDR1 to 3 of SEQ ID: Nos 1-3, and L-CDR of SEQ ID NOs: 4-6, or (ii) a heavy chain of SEQ ID NO: 7 and a light chain of SEQ ID NO: 8; (b) 20 mM histidine, (c) 75 mM of arginine hydrochloride, (d) 150 mM sucrose, (e) 0.5 mg / mL PS80 and (f) 0.2 mM of EDTA, at pH 5.85. In some aspects, the composition is about 20, 50 or lOOmg / ml.

[0097] In some aspects, the pharmaceutical composition comprises: (a) about 150 mg / mL of TEV-53408 that comprises (i) H-CDR1 to 3 of SEQ ID: Nos 1-3, and L-CDR of SEQ ID NOs: 4-6, or (ii) a heavy chain of SEQ ID NO: 7 and a light chain of SEQ ID NO: 8; (b) 20 mM histidine, (c) 180 mM of arginine hydrochloride, (d) 180 mM sucrose, (e) 0.5 mg / mL PS80 and (f) 0.1 mM of EDTA, at pH 5.2.

[0098] In some aspects, the pharmaceutical composition comprises antibody diluted to less than 100 mg / mL, to enable administration of lower doses such as to 1, 2, 5, 7.5, 15, 20, 25, 35, 40, 45 and 50 mg, in a volume that is not so small as to introduce dosage error. For example, many autoinjectors require at least 0.4 ml, such that a lOOmg / mL pharmaceutical composition is typically diluted to administer a dose less than 40mg. For immediate use, the composition can be diluted in an isotonic dextrose solution. For longer term storage, the composition can be diluted in the samebuffered composition. Accordingly, the composition can comprise 20 to 50 mg / mL antibody, (b) 20 mM histidine, (c) 250 mM sucrose, (e) 0.5 mg / mL PS80 and (f) 0.1 mM of EDTA, at pH 5.2.

[0099] In some aspects, the pharmaceutical formulations provided herein have no significant difference in stability after storage at 2-8° C for up to 24 months, 25°C for up to 18 months, or 40°C for up to 6 months.

[0100] In some aspects, the pharmaceutical formulations provided herein have at least about 98% antibody monomer content after storage at 25 °C for up to 3 months.

[0101] In some aspects, the pharmaceutical formulations provided herein have no significant difference in viscosity, hydrodynamic radius, or polydispersity after storage at 2-8°C for up to 24 months.

[0102] In some aspects, the pharmaceutical formulations provided herein have no significant difference in sub-visible particle content after storage at 2-8°C for up to 24 months, 25°C for up to 18 months, or 40°C for up to 6 months.

[0103] In some aspects, the pharmaceutical formulations provided herein have from about 60% to about 140% relative potency measured by a cell-based potency assay after storage at room temperature for 24 hours, and after storage at 40°C for up to 1 month.

[0104] In some aspects, the pharmaceutical formulations provided herein have no significant difference in thermal stability after storage at 2-8°C for up to 24 months or 25°C for up to 6 months.

[0105] In some aspects, the pharmaceutical formulations provided herein have no significant difference in thermal stability after storage at 2-8°C for up to 24 months.

[0106] In some aspects, the pharmaceutical formulations provided herein have no significant difference in secondary and / or tertiary protein structure after storage at 2-8°C for up to 24 months or 25°C for up to 18 months.

[0107] In some aspects, the pharmaceutical formulations provided herein have no significant difference in secondary protein structure after storage at 2-8°C for up to 24 months.Therapeutic Methods and Doses

[0108] The disclosure provides methods for treating an immune disease in a subject in need thereof by administering to the subject the pharmaceutical formulation of TEV-53408 antibody provided herein. The disclosure also provides methods for treating an immune disease by administering TEV-53408 antibody to a person in need thereof at the doses and frequencies provided herein. The appropriate dose and frequency can vary by condition and by patient.Typically, frequency is once a month or once every 3-6 months. A frequency of once a month is understood to be once every calendar month, every 4 weeks, or every 25-35 days, particularly about 28-30 days, more particularly every 28 days. A 3 month dosage schedule can be calculated as 3 calendar months, 12 weeks, or about 90 days.

[0109] A dose can range from about 1 mg to about 700 mg. In some aspects, a dose is about 5 mg. In some aspects, a dose is about 5 mg, about 7.5 mg, about 10 mg, about 20 mg, about 25 mg, about 30 mg, or about 50 mg administered once a month. Preferably, a dose is about 10, about 20, about 30, or about 50 mg / month. In some aspects, a dose is about 50 mg to about 700 mg, preferably about 50 mg to about 350 mg, or preferably about 50 mg to about 150 mg every 3 months. While 150 mg every 3 months has the same total dose as 50 mg per month, the pharmacokinetics and pharmacodynamics can be different.

[0110] Less frequent dosage is possible. Free serum IL- 15 levels were suppressed below baseline level for more than 200 days with 150mg (Figure 18) and more than 90 days with 50mg (Figure 17). 50 and 150 mg was able to suppress NK-cell levels for over 190 days, which is consistent with that a single dose of antibody is able to suppress IL- 15 activity for a period beyond 3 months, such that a therapeutic effect can be obtained by dosing every 4-6 months at about 50 mg to about 700 mg, preferably 100 mg to about 150 mg.[OHl] The appropriate dose can be determined by a combination of pharmacokinetics (measuring levels of antibody) and pharmacodynamics (measuring IL- 15 and / or the effect of the drug, such as on NK cells), with the ultimate goal being treatment of disease.

[0112] NK cell levels can be measured directly by cell counts. They can also be measured indirectly by one or more biomarkers.

[0113] Effectiveness has been demonstrated in animal models, including celiac disease, vitiligo, alopecia and atopic dermatitis. In each of these, effectiveness in treatment or prevention of disease pathology was accompanied by suppression of CD8+ Trm cells, suppression of various inflammatory mediators and cells, suppression of free circulating IL- 15, and suppression of circulating NK cells. Thus, even if circulating IL-15 and circulating NK cells are merely correlates, they can serve as pD guides for a dose sufficient to treat disease. According, serum NK cell and IL- 15 levels can serve as indicators for a clinically relevant dose.

[0114] Should a higher dose be required, the antibody is well tolerated. There were no serious adverse events in mice at 2.5mg / kg or monkeys up to 400mg / kg. These doses are equivalent to 175mg and 28,000 mg, respectively, in humans. The highest dose tested in humans so far is 150mg, and did not cause serious treatment-related adverse events. There was no observedelevation in infection or cancer, despite reduction of serum IL- 15 to BLQ, and a decrease in NK cells by approximately 70%, and persistent NK suppression for over 6 months.Treatment of celiac disease

[0115] The treatment of celiac disease can be determined by different endpoints. Attenuation of gluten-induced small intestinal mucosal injury as measured by the villus height / crypt depth (“V / C”) ratio is considered by some to be the “gold standard” for clinical trials, but can not be appropriate for treatment endpoints in the general population, not least because of the requirement for biopsy, and can not be fully reflected in other markers of disease. Attenuation of gluten-induced small intestinal mucosal inflammation as measured by the enumeration of intraepithelial lymphocytes (lELs) in histological sections is also relevant. Attenuation of gluten- induced serum antibodies such as anti-gliadin antibodies and autoantibodies against transglutaminase can indicate a reduction in inflammatory processes. Other reduction in inflammatory markers, or inflammatory conditions secondary to celiac disease, can also be monitored. For example reduction in Th2 associated cytokines; activated CD8 cells, and the like. Patient reported outcomes, as assessed through questionnaires, can capture the broader symptomatic reduction and subject experience.

[0116] Fatty acid-binding protein 2 (FABP2), also known as Intestinal -type fatty acidbinding protein (LFABP), is an enterocyte protein that is released in the circulation when gut integrity is disrupted. Serum I-FABP has been observed to correlate to enterocyte damage in coeliac disease and its relation to villous atrophy and circulating autoantibodies (Adriaanse et al. Aliment Pharmacol Ther. 2013 Feb;37(4):482-90). Serum I-FABP levels increased significantly during a two-week gluten challenge in adult CD patients and correlated with IEL count, but did not correlate with the gluten dose nor time on a GFD. (Adriaanse et al Am J Gastroenterol. 2016 Jul; 111(7): 1014-22). Hoffmanova et al (Physiol Res. 2015;64(4):537-46.) also showed that celiac disease patients have higher levels of I-FABP, that the level was reduced on gluten free diet but failed to return to control levels. I-FABP was also raised in T1D and T2D. Accordingly, we propose that I-FABP levels can be monitored to determine gut health, including the ability of TEV- 53408 to prevent gluten-mediated enteropathy, and treatment of enteropathy whether on a gluten- free diet or with gluten exposure.

[0117] Provided herein are methods for repairing the mucosa of a small intestine in a subject having celiac disease; for increasing the mean villous height vs. crypt depths (V / C) ratio in a subject having celiac disease; and / or decreasing the symptoms of celiac disease as assessedthrough patient reported outcomes; comprising administering to the subject a pharmaceutical formulation containing TEV-53408. Suitable pharmaceutical formulations containing TEV-53408 are described herein.

[0118] In some aspects, the celiac disease is refractory. Refractory celiac disease (RCD) affects patients who have failed to heal and demonstrate continual symptoms of celiac disease, after 6-12 months of a strict gluten-free diet and when other causes of symptoms (including malignancy) have been ruled out. It can also occur in patients who previously responded to a longterm gluten-free diet, but who now display symptoms of celiac disease, while maintaining a strict gluten free diet (Rishi et al. Expert Review of Gastroenterology & Hepatology: 10 537-546 (2016)).

[0119] Also provided herein are methods for inhibiting one or more symptoms of gluten exposure in a subject having a gluten sensitivity or gluten allergy and in need thereof, comprising administering to the subject TEV-53408 or pharmaceutical formulation as described herein.

[0120] The one or more symptoms of gluten exposure can include one or more of muscle pain, body pain, joint pain, fatigue, bloating, gas, nausea, cramps, constipation, diarrhea, skin rash, headache, migraine headache, depression, anxiety, brain fog, and / or irritability. See, Biesiekierski JR, United European Gastroenterol. J. 3: 160-165 (2015).

[0121] In some aspects, the subject consumes a gluten-containing diet.

[0122] In some aspects, the pharmaceutical formulations as described herein are administered subcutaneously.Treatment of vitiligo

[0123] Vitiligo is a common (0.5-2%) skin depigmentation disorder, characterized by patches of skin losing its pigment. Although sometimes regarded as merely a cosmetic disease; it can have a significant impact on the quality-of-life of individuals due to stigmatization, social isolation, low self-esteem, and mental health disorders. Current treatments include UV phototherapy, systemic steroids and other immunosuppressants, and local anti-inflammatory creams, and JAK inhibitors. Each of these treatments have problems including poor compliance, serious side effects and, upon cessation of treatment, vitiligo reoccurs in 40% of the cases (Frisoli et al 2020). Vitiligo is caused by cytotoxic T lymphocytes targeting melanocytes, and melanocytespecific CD8+ resident memory T cell (TRM) constitute a reservoir for cytotoxic T cells. Keratinocytes in lesional vitiligo skin express elevated levels of IL-15Ra and CD8+ T cells exposed to IL- 15 in trans presentation will elevate expression of CD69, CD 103, and CD49a, which are markers for activity and residence within the tissue. These cells also over express CXCR3 athigh levels, enabling their recruitment to the skin by CXCL9 / 10-dependent chemotaxis. Upon stimulation, CD8+ TRMs produce IFN-y and TNF-a, which mediates their anti-melanocyte cytotoxicity and CXCL9 and CXCL10, promoting recruitment of circulating memory T cells to the skin.

[0124] Treatment of vitiligo can be monitored by calculating the Facial Vitiligo Area Scoring Index (F-VASI) and / or the Total Vitiligo Area Scoring Index (T-VASI), typically by an expert physician. Cytokine markers and T-cells can be measured in skin via suction blister or punch biopsy.Alopecia areata

[0125] Alopecia areata occurs when the immune system attacks the hair follicle leading to patchy loss of hair. Disease progression can be measured by hair loss and growth, such as the amount of hair, its thickness and coverage, as well as by immune markers, such as CD4+T cells, CD8+NKG2D+T cells, and IFNy. Accordingly, treatment with TEV-53408 can be observed by attenuation of hair loss, increase in hair growth, and reduction in immune markers.Atopic dermatitis

[0126] Atopic Dermatitis (AD) is a chronic inflammation of the skin, manifested as dry, itchy and inflamed skin. AD is mainly driven by a type 2 helper T cell (Th2) response, but eosinophils and CD8+ T cells are also involved. AD progression and treatment can be measured symptomatically including erythema (redness), lichenification (crusting over), itch, skin thickening, immune infiltrates and skin barrier breach. Immune measurements include the presence and activity of immune cells and inflammatory cytokines, including CD4+GATA3+T cells, CD8+GATA3+T cells, IL-4, IL-13, TSLP and more.Kits and Containers

[0127] The disclosure also features containers comprising any pharmaceutical formulation described and exemplified herein. In some aspects, the container is a glass vial.

[0128] In some aspects, the container is a syringe. In some aspects, the syringe is a prefilled syringe. In some aspects, the syringe is a glass, polypropylene or polycarbonate syringe.

[0129] In some aspects, the container is an autoinjector.Analytical methods

[0130] The disclosure also provides methods of measuring pharmacokinetics and pharmacodynamics.Embodiments

[0131] The invention can be further understood by reference to one or more embodiments.

[0132] Embodiment 1. An aqueous pharmaceutical formulation, comprising:(a) about 100 mg / mL to about 150 mg / mL of an antibody that specifically binds to human IL-15, wherein the antibody or antigen-binding fragment thereof comprises: a heavy chain variable region complementarity determining region (CDR)l comprising the amino acid sequence of SEQ ID NO: 1, a heavy chain variable region CDR2 comprising the amino acid sequence of SEQ ID NO: 2, a heavy chain variable region CDR3 comprising the amino acid sequence of SEQ ID NO: 3, a light chain variable region CDR1 comprising the amino acid sequence of SEQ ID NO: 4, a light chain variable region CDR2 comprising the amino acid sequence of SEQ ID NO: 5, and a light chain variable region CDR3 comprising the amino acid sequence of SEQ ID NO: 6.; and(b) one or more pharmaceutically acceptable excipients.

[0133] Embodiment 2. The aqueous pharmaceutical formulation of embodiment 1, wherein the antibody comprises a heavy chain variable region comprising the amino acid sequence of SEQ ID NO: 7 and a light chain variable region comprising the amino acid sequence of SEQ ID NO: 8.

[0134] Embodiment s. The aqueous pharmaceutical formulation of any foregoing embodiment, wherein the antibody or antigen-binding fragment comprises a human IgG4 heavy chain constant region comprises an S228P substitution (by EU numbering), a triple substitution of M252Y / S254T / T256E (by EU numbering), and terminal lysine deletion (K447A) (by EU numbering).

[0135] Embodiment 4. The aqueous pharmaceutical formulation of any foregoing embodiment, wherein the antibody or antigen-binding fragment further comprises a human lambda light chain constant region.

[0136] Embodiment 5. The aqueous pharmaceutical formulation of any foregoing embodiment, wherein the antibody comprises a heavy chain comprising the amino acid sequence of SEQ ID NO: 9 and a light chain comprising the amino acid sequence of SEQ ID NO: 10.

[0137] Embodiment 7. The aqueous pharmaceutical formulation of any foregoing embodiment, comprising about 20, about 50 or about 100 mg / mL of the antibody.

[0138] Embodiment 8. The aqueous pharmaceutical formulation of any foregoing embodiment, comprising about 150 mg / mL of the antibody.

[0139] Embodiment 9. The aqueous pharmaceutical formulation of any foregoing embodiment wherein said pharmaceutically acceptable excipients include histidine, sucrose, EDTA and polysorbate.

[0140] Embodiment 10. The aqueous pharmaceutical formulation of embodiment 9, comprising (b) about 10 mM to about 50 mM Histidine; (c) about 150 mM to about 350 mM Sucrose; (d) about 0.01 mM to about 0.5 mM EDTA; and (e) about 0.1 mg / mL to about 0.9 mg / mL Polysorbate-80.

[0141] Embodiment 11. The aqueous pharmaceutical formulation of embodiment 10, comprising about 20 mM Histidine.

[0142] Embodiment 12. The aqueous pharmaceutical formulation of embodiment 10, comprising about 150 mM Sucrose.

[0143] Embodiment 13. The aqueous pharmaceutical formulation of embodiment 10, comprising about 180 mM Sucrose.

[0144] Embodiment 14. The aqueous pharmaceutical formulation of embodiment 10, comprising about 250 mM Sucrose.

[0145] Embodiment 15. The aqueous pharmaceutical formulation of embodiment10,, comprising about 0.1 mM EDTA.

[0146] Embodiment 16. The aqueous pharmaceutical formulation of embodiment 10, comprising about 0.5 mg / mL Polysorbate-80.

[0147] Embodiment 17. The aqueous pharmaceutical formulation of any foregoing embodiment, further comprising about 75 mM to about 180 mM Arginine-Hydrochloride (Arg- HC1).

[0148] Embodiment 18. The aqueous pharmaceutical formulation of any foregoing embodiment, comprising about 100 mg / mL of the antibody, about 20 mM Histidine, about 250 mM Sucrose, about 0.1 mM of EDTA, and about 0.5 mg / mL Polysorbate-80.

[0149] Embodiment 19. The aqueous pharmaceutical formulation of any foregoing embodiment having a pH of 5.2 ± 0.5.

[0150] Embodiment 20. The aqueous pharmaceutical formulation of any foregoing embodiment having an osmolality of from about 200 mOsm / kg to about 350 mOsm / kg at room temperature.

[0151] Embodiment 21. The aqueous pharmaceutical formulation of any foregoing embodiment, having no significant difference in stability after storage at 2-8° C for up to 24 months, 25°C for up to 18 months, or 40°C for up to 6 months.

[0152] Embodiment 22. The aqueous pharmaceutical formulation of any foregoing embodiment having: at least about 98% antibody monomer content after storage at 25 °C for up to 3 months; having no significant difference in viscosity, hydrodynamic radius, or polydispersity after storage at 2-8°C for up to 24 months; having no significant difference in sub-visible particle content after storage at 2-8°C for up to 24 months, 25°C for up to 18 months, or 40°C for up to 6 months; having from about 60% to about 140% relative potency measured by a cell-based potency assay after storage at room temperature for 24 hours, and after storage at 40°C for up to 1 month; having no significant difference in thermal stability after storage at 2-8°C for up to 24 months or 25°C for up to 6 months; having no significant difference in secondary and / or tertiary protein structure after storage at 2-8°C for up to 24 months or 25°C for up to 18 months; and / or having no significant difference in secondary protein structure after storage at 2-8°C for up to 24 months.

[0153] Embodiment 23. A container comprising the aqueous pharmaceutical formulation of any foregoing embodimen

[0154] Embodiment 24. The container of embodiment 23, wherein the container is selected from (a) a glass vial; (b) a pre-filled syringe; (c) a polyolefin or polyvinyl chloride bag; and (d) an autoinjector.

[0155] Embodiment 25. A method of treating an immune disease in a subject in need thereof, comprising administering to the subject an antibody comprising HCDR 1-3 comprising the amino acid sequences of SEQ ID NOs: 1-3, respectively, and LCDR 1-3 comprising the amino acid sequences of SEQ ID NOs: 4-6, respectively, at a dose from about 10 to about 700 mg.

[0156] Embodiment 26. The method of embodiment 25, wherein dose is from about10- about 50 mg every 4 weeks, or about 50- about 150mg every 12 weeks.

[0157] Embodiment 27. The method of embodiment 26, wherein the dose is selected from about lOmg every 4 weeks, about 20 mg every 4 weeks, about 30 mg every 4 weeks, about 50 mg every 4 weeks, about 50 mg every 12 weeks, and about 150 mg every 12 weeks.

[0158] Embodiment 28. The method of embodiment 25, wherein the immune disease is an autoimmune disease

[0159] Embodiment 29. The method of embodiment 28, wherein the autoimmune disease is selected from celiac disease, vitiligo, alopecia, atopic dermatitis, eosinophilic esophagitis, Sjogrens syndrome, graft-versus host disease, type I diabetes, myositis, and rheumatoid arthritis.

[0160] Embodiment 30. The method of embodiment 28, wherein the treatment of autoimmune disease is characterized by (a) a reduction in the number of activated CD8 T resident memory cells and effector cells at a site of disease, (b) reduction in circulating CD8 T memory cells (c) reduction in inflammatory cytokines and / or (d) reduction in pathology.

[0161] Embodiment 31. The method of embodiment 30 wherein the reduction in inflammatory cytokines is reduction in any one of IFND, TNFD, IL-4, IL-13, IL-31, TSLP and / or TARC.

[0162] Embodiment 32. The method of embodiment 31 wherein the reduction in effector cells is reduction in CD4+ and / or CD8+ T cells.

[0163] Embodiment 33. The method of embodiment 25, wherein the method of treating an immune disease in a subject in need thereof comprising administering to the subject does not cause serious adverse effects.

[0164] Embodiment 34. The method of embodiment 33, wherein the administering to the subject does not cause a decline in circulating NK cell numbers below 1% of total lymphocytes; does not significantly affect the CD56bright and %CD56dim ratio in circulating NK cells and / or does not significantly affect NK cell function in remaining NK cells.

[0165] Embodiment 35. The method of embodiment 34, wherein NK cell function in remaining NK cells is assessed by CD 107a expression or perforin expression.

[0166] Embodiment 36. The method of embodiment 33, wherein administering to the subject inhibits NK cell maturation to stage 4 in the bone marrow, but does not affect NK cell progenitors.

[0167] Embodiment 37. The method of embodiment 29, for the treatment of celiac disease, wherein the treatment of celiac disease in the subject comprises: (a) repairing the mucosa of a small intestine; (b) increasing the mean villous height vs. crypt depths (V / C); and / or (c) decreasing symptoms of one or more of muscle pain, body pain, joint pain, fatigue, bloating, gas, nausea, cramps, constipation, diarrhea, skin rash, headache, migraine headache, depression, anxiety, brain fog, and irritability; as assessed by patient reported outcomes.

[0168] Embodiment 38. The method of embodiment 29, for the treatment of vitiligo, wherein the treatment of vitiligo in the subject comprises (a) reduction in the rate of skindepigmentation, (b) increase in the area of skin pigmentation (c) reduction in the number and / or activity of melanocyte-targeting T-cells in the skin and / or (d) reduction in the number and / or activity of melanocyte-targeting T-cells in other organs.

[0169] Embodiment 39. The method of embodiment 38, wherein skin pigmentation is measured by F-VASI and / or T-VASI.

[0170] Embodiment 40. The method of embodiment 38, wherein the subj ect is further treated with ultraviolet light.

[0171] Embodiment 41. The method of embodiment 29 for the treatment of alopecia areata, wherein the treatment of alopecia areata in the subject comprises (a) reduction in the rate of hair loss, (b) increase in hair (c) reduction in the number and / or activity of hair follicle-targeting T-cells in the skin and / or (d) reduction in the number and / or activity of melanocyte-targeting T- cells in other organs.

[0172] Embodiment 42. The method of embodiment 29 for the treatment of atopic dermatitis, wherein the treatment of atopic dermatitis comprises (a) reduction of erythema (b) reduction in lichenification (c) reduction in skin thickening (d) reduction in hyperproliferation and expansion of epidermal cells (as evidenced by, e.g., Ki67 and Krtl6 staining) (e) increased expression of filaggrin (f) reduction in transepithelial water loss (TEWL) (g) reduction the levels of CD4+ and CD8+ T cells in the skin and / or (h) suppression of IL-4, IL-13, IL-31, TSLP and / or TARC.ExamplesExample 1: Effects of pH and Excipients on TEV-53408 Drug Product StabilityZeta Potential

[0173] TEV-53408 in 25 mM Histidine buffer at pH 6.17 was titrated from its original pH6.17 to pH 8.7 with sodium hydroxide while measuring the Zeta potential (mV) and the average hydrodynamic diameter (nm) by dynamic light scattering (DLS) (Figure 1). The isoelectric point , was determined as 7.83 (potential = 0). Higher colloidal stability of TEV-53408 is expected with , potential > + 3.0 mV, i.e., a pH more acidic than 6.17. g. The , -average (Figure 1) also decreases from 7.2 to 6.17, suggesting higher colloidal stability at pH <6.6. Both , potential and DLS stability results suggests that TEV-53408 should be more stable at pH at or below 6.17.Test formulations

[0174] To evaluate stability and compatibility with excipients over a pH range from 5.5 to 6.5, antibody was formulated in a histidine buffer system with three osmolytes (arginine, sucrose, and sodium chloride). TEV-53408 drug substance (DS) at 123.4 mg / mL in 25 mM Histidine at pH 6.2 was dialyzed and spiked into buffer stock solutions to arrive at the final concentrations inTable 1Table 1: Excipient and pH Combinations Used in TEV-53408 Stability Study

[0175] Stability was evaluated by visual appearance, pH, sub visible particle count by micro-flow imaging (MFI), protein concentration (A280), turbidity (A350 and A450), size exclusion chromatography (SEC), non-reducing capillary sodium dodecyl sulfate electrophoresis (CGE), imaged capillary isoelectric focusing (icIEF), intrinsic fluorescence, dynamic light scattering (DLS), and viscosity measurements. This was immediately after preparation (TO) and after 1 month of storage at +40 ° C temperature (IM at 40°C).Visual Appearance, Turbidity, Hydrodynamic Properties, and Colloidal Stability by DLS

[0176] The appearance of each formulation is summarized in Table 2. All samples had visible particles. It cannot be conclusively determined whether these are results of formulationprotein interactions or environmental contaminants because the experimental samples were notprepared in a GMP -validated dust free environment, and because bubble formation interferes with visible particle inspection. The number of particles did not increase after 40°C stress.Table 2: Visual inspection and turbidimetric measurements show effect of stress.TO: initial time point; 1M40C: after 1 month storage at 40°C; O: opalescent; S: slightly opalescent; HO: highly opalescent

[0177] At TO, the formulations were liquid clear opalescent solutions, and A02, A08, A09, and Al 1 were noticeably less opalescent. A03 exhibited phase separation that disappeared upon swirling. After 1 month at 40°C A01, A02, A04, A05, A06, A07, and A10 were more opalescent than A08, A09, and Al l. The solution in the formulation A03 was the most opalescent of all formulations.

[0178] Turbidity was evaluated using UV-Vis absorbance at 400 nm, 450 nm, and 500 nm against a buffer blank, with 20 nephelometric turbidity unit (NTU) and 200 NTU standards. The measured absorbances correlated with the visual inspection results shown in Table 2. Statistical analysis of the turbidity measurements indicated that a formulation with 40 mM histidine, 150 mM sucrose at pH 5.5 has the lowest absorbance at 400 nm, and as a result, the highest colloidal stability by A400.

[0179] TEV-53408 diffusion coefficient as a function of protein concentration (1.0, 2.0,4.0, 8.0, 15.0, 25.0 and 50.0 mg / mL) as measured by DLS is shown in Table 3. The diffusioninteraction parameter kD and protein self-diffusion coefficient at infinite dilution DO were obtained from the least square linear fits (Table 3). In formulations with sucrose the 1.0 mg / mL and in some cases the 2.0 mg / mL values were excluded from the fit due to DLS sensitivity to concentrated sucrose solutions. The viscosity of the TEV-53408 formulations is summarized in Table 3. All formulations at 100 mg / mL protein concentration showed acceptable viscosities ranging from 4.0 cP to 12.1 cP.Table 3: Diffusion Interaction Parameter (kD), Self-Diffusion Coefficient at Infinite Dilution (DO) and viscosity in TEV-53408 formulations

[0180] Table 4 shows TEV-53408 colloidal stability evaluated using the average hydrodynamic radius (Rh) of the molecule and the polydispersity (%Pd) at the target concentration of 100 mg / mL. The Rh and %Pd increased after 1 month storage at 40°C as compared to initialtime point. The increase in Rh was minor for all formulations, with the exception of F02, F03, and F10. These three formulations showed the increase in Rh by more than 5 nm. The polydispersity (%Pd) increased for all formulations except A03. Higher Pd% indicates broader distribution in hydrodynamic radius of the protein molecules which can indicate aggregate formation.Table 4: Hydrodynamic radius (Rh) and polydispersity (%Pd) measured at 100 mg / ml protein concentration of TEV-53408.Note: DS = drug substance: 123.4 mg / mL TEV-53408 in 25 mM Histidine at pH 6.2

[0181] Statistical analysis of the DLS and viscosity measurements indicated that a formulation with 40 mM histidine, 150 mM sucrose, and 150 mM arginine at pH 5.5 is expected to have higher colloidal stability and lower viscosity.

[0182] The aggregation propensity of TEV-53408 in formulations was explored by dynamic light scattering (DLS) with temperature scan (Figure 2A), and the hydrodynamic radius (Rh) and polydispersity (%Pd) were used to track changes in protein size and distribution during temperature scan from 25°C to 72.5°C (Figure 2B). The aggregation onset temperatures are shown in Table 5. Two thermal transitions in the hydrodynamic radius (Rh) thermograms were observed for all formulations, except A01, A02, and A03. Formulations A01 and A02 showed higher thermal stability with Tonset 1 for the first transition higher than 65°C, and with no clear increase in Rh. The polydispersity (%Pd) thermograms all showed two transitions and slightly lower Tonsettemperatures for all formulations. Formulations A01 and A02 were also more thermally stable than the rest. The lowest thermal stability was observed for formulations A09 and A10 with a Tonset at 50°C.Table 5: Thermal aggregation parameters of TEV-53408 by DLS

[0183] Statistical analysis of the Rh and %Pd measurements indicated that a formulation with 40 mM histidine and 150 mM sucrose at pH 6.7 was expected to have higher thermal stability and lower aggregation propensity.Sub-Visible Particulate Matter by MFI

[0184] Table 6 and Table 7 shows the concentration of sub-visible particles by MFI for TO and after 1 month of storage at 40°C. There was a significant increase in sub-visible particles in all formulations after sample storage at 40°C.Table 6: Concentration of sub-visible particles by MFI in TEV-53408 formulations at the initial time pointTable 7: Concentration of sub-visible particulate per mL by MFI in TEV-53408 formulation after 1 month of storage at 40°CPurity by Size Exclusion Chromatography (SE-HPLC)

[0185] Purity as evaluated by size exclusion high-performance liquid chromatography (SE- HPLC), is shown in Table 8. Heat stress (1M@4O°C) caused an increase in low molecular weight species (LMW) and significant dimer formation. Higher order aggregates (HOA) were detected after IM at 40°C only in formulations A02 and A03.Table 8: Purity of TEV-53408 by SE-HPLC at initial time and after 1 month at 40°CNote: DS = drug substance; TO = initial time point; NT = not tested; ND = not detectedPurity by non-reducing capillary gel electrophoresis (CGE)

[0186] Purity as assessed by non-reducing capillary gel electrophoresis (CGE) is shown in Table 9. There were no significant differences between formulations at the initial time point. After 1 month at 40°C an increase in fragments of 0.8% - 1.1% was observed across all formulations.Table 9: Purity of TEV-53408 by non-reducing CGENote: DS = drug substance; TO = initial time point; NT = not testedPurity by imaged capillary iso-electric focusion (icIEF)

[0187] Charge heterogeneity of the TEV-53408 formulations was determined by imaged capillary iso-electric focusing (icIEF). As shown in Table 10, the distribution of main charge variant, total acidic variants and total basic variants between DS and the formulations at TO did notsignificantly vary, and the small differences could be attributed to method variability. 1 month storage at 40°C did not significantly affect the basic isoforms, but the main isoform decreased by 16.9 % - 21.1 %, a similar increase in acidic isoforms.Table 10: Charge Heterogeneity in TEV-53408 FormulationsNote: DS = drug substance; TO = initial time point; NT = not tested

[0188] A statistical evaluation of the icIEF measurements indicated that formulation pH had the most significant impact on charged isoforms after IM storage at 40°C. A formulation with 12.5 mM Histidine, 150 mM Arginine HC1, 40 mM NaCl, and 150 mM Sucrose at pH 5.6 was expected to have more main and less acidic charge variants as measured by icIEF.Global Ranking of TEV-53408 Formulation Variables

[0189] Table 11 shows a statistical data analysis for each quality attribute (histidine, sodium chloride, sucrose, arginine concentration and solution pH) separately to identify optimal formulation compositions. Each quality attribute was assigned a ranking number 0 to 1 as with the sum of all rankings equal to 1. The global fit of formulation parameters was then performed using JMP prediction profiler to identify global fit and prediction profile for all parameters with maximized desirability (Figure 3), identifies an optimal formulation for TEV-53408 at lOOmg / ml in 37 mM histidine, 75 mM of arginine hydrochloride, 150 mM sucrose at pH 5.85. To reduce thetheoretical osmolality of the final formulation to better match tissue osmolality, the histidine concentration was reduced to 20 mM. No sodium chloride is preferred in the formulation.Table 11: Prediction Profile Responses for Maximized Desirability of TEV-53408 Quality AttributesExample 2: Effects of Surfactant and Antioxidant on TEV-53408 Drug Product Stability

[0190] The purpose of this study was to evaluate the effect of the surfactant polysorbate 80 (PS80) and the anti oxi dant / chelator ethylenediaminetetraacetic acid (EDTA) on the stability of a formulation of 100 mg / mL TEV-53408 in 20 mM histidine, 75 mM arginine HC1, and lOOmM sucrose at pH 5.85.

[0191] The formulations tested for the surfactant and antioxidant study are shown in Table 12Table 12: Addition of polysorbate and EDTA to 100 mg / mL TEV-53408 in 20 mM histidine, 75 mM arginine HC1, and lOOmM sucrose at pH 5.85, in glass vials.

[0192] In all samples at the initial time point after vial filling, the measured protein was 100 ± 4 mg / mL and the measured pH was 5.85 ± 0.1.

[0193] The remaining vials were set aside for testing under the stressed conditions as shown in Table 13. The spiking with Fe3+was performed by the addition of 150 ppm Fe3+stock into a 6R vial with TEV-53408 formulation, to a final concentration of 1 ppm (1.8nM), then incubated at 40°C for 2 weeks.Table 13: Stress Conditions Evaluated for TEV-53408 Formulations

[0194] Pulled samples were analyzed immediately for appearance, visible particulate matter and clarity and degree of opalescence, for sub-visible particulate matter by MFI, for protein concentration by slope spectroscopy with variable path length (SoloVPE), and for solution pH. Samples for the remaining analytical tests were aliquoted and stored frozen at <-65°C until analysis.Visual Appearance, Turbidity, Hydrodynamic Properties, and Colloidal Stability by PLS

[0195] All formulations were opalescent. Agitation and light did not affect opalescence. B01-B04 were particularly sensitive to stirring and 3FT. These contain 0.1 mg / mL polysorbate 80, suggesting that higher than 0.1 mg / mL concentration can be required for stabilization of TEV- 53408. Turbidity measurements UV-Vis absorbance at 400 nm, against formulation buffer and 20 NTU and 200 NTU and stirring showed all formulations were opalescent. Opalescence wasincreased by stirring, but not by the other stressors. The conclusion is that higher PS 80 concentrations can result in lower turbidity.

[0196] DLS analysis showed that the hydrodynamic radius (Rh) at TO was 13.9-14.3 nm in all formulations and no significant increase was observed after agitation, light exposure, or after 3 FT cycles. Stirring increased the hydrodynamic radius by more than 4 nm and the polydispersity index (%PD) increased more than threefold. All formulations at 100 mg / mL protein concentration showed an acceptable viscosity around 4.9 cP. No significant impact of PS80 and EDTA concentrations was observed on hydrodynamic radius and polydispersity by DLS, or on solution viscosity, across the formulations.

[0197] TEV-53408 aggregation was monitored by DLS during a temperature scan from25°C to 72.5°C in 2.5°C increments. The associated hydrodynamic radius (Rh) and polydispersity (%Pd) data are summarized in Table 14. The hydrodynamic radius and the polydispersity index as a function of temperature showed a thermal transition at 52.5-55°C and 62.5-65°C, but no significant differences across the TEV-53408 formulations.Table 14: Aggregation onset (Tonset 1 and Tonset 2) and aggregation (Taggr) temperatures and the polydispersity index onset (%PD Onset 1) of TEV-53408 by DLSSub-Visible Particles by Micro-Flow Imaging (MFI)

[0198] The sub-visible particle concentration was similar in each formulation at TO, and was largely unaffected by agitation, freeze-thaw, and light stress. Stirring had a strong impact on sub-visible particle concentration. Fewer sub-visible particles were detected in formulations with 0.5 mg / ml and 0.9 mg / mL PS80 concentration than in formulations with 0.1 mg / mL PS80. Thesesub-visible particle results correlate well with the hydrodynamic radius results determined by DLS. PS80 concentration was also statistically significant in reducing the concentration of 1 pm-10 pm sub-visible particles in initial samples, and in response to stirring stress. EDTA concentration was statistically significant with regards to reducing 10pm-25pm size sub-visible particles, and was effective during light stress. Based on statistical analysis of the sub-visible particle data, an optimal formulation composition was found to be 0.9 mg / mL of PS80 and 0.18 mM of EDTA based on MFI data.Purity of TEV-53408 by Size Exclusion Chromatography (SE-HPLC)

[0199] Purity as determined by SE-HPLC at TO showed 96.9% monomer, 3.0-3.1% dimer, 0.07-0.08% LMW and no detectable HOA across all formulations. Purity by SE-HPLC was not significantly impacted by stressors except light, which decreased monomer and increased dimer. The prediction profiler for TEV-53408 purity by SE-HPLC did not show PS80 or EDTA concentrations having significant impact (results not shown).Purity of TEV-53408 by Non-Reducing Capillary Gel Electrophoresis (CGE)

[0200] Purity as assessed by non-reducing capillary gel electrophoresis (CGE) at TO was 98.6%-98.7% IgG, 0.6%-0.7% of fragments, and 0.7%-0.85 of HHL truncated species across all formulations. No significant changes were observed in any of the formulations under agitation, stirring, or light stress. 3FT samples could not be not tested by non-reducing CGE due to high solution opalescence. (Results not shown)Charge Heterogeneity of TEV-53408 by Imaged Capillary Iso-Electric Focusing (icIEF)

[0201] The charged species distribution in the TEV-53408 formulation samples at TO was 69.7%-71.1% of main isoform, 22.9%-24.0% of acidic isoforms, and 5.8%-6.4% of basic isoforms. No significant change was observed after agitation or stirring stress. Light caused h caused an approximately 1% loss of main isoform and concomitant 1% increase in the acidic isoforms, with no significant change in basic isoforms. These results were observed across all formulations with no significant formulation based differences.Peptide Mapping Analysis of Selected TEV-53408 Formulations

[0202] To identify any possible impact of PS80 and EDTA on chemical modifications of TEV-53408, peptide mapping was performed on formulations B01, B03, B05, B07 and B09 at TOand after spiking with Fe+3two-weeks of storage at 40°C (Table 15). Tryptophan at position 103 (W103) in the complementarity-determining region (CDR) is considered highly solvent accessible and its oxidation has been determined to have an impact on TEV-53408 function / potency (data not shown).

[0203] Methionines M357 and M427 were most frequently oxidized. A small increase in oxidation after stress was also detected for tryptophan W 103 in TEV-53408 formulation B07. The data in Table 15 show that TEV-53408 in formulation B07 is the most susceptible to oxidation. These results suggest that the combination of high PS80 (0.9 mg / mL) and low EDTA (0 mM) in the presence of histidine is most likely to promote oxidation, and that EDTA should be included in the formulation to reduce the possibility of oxidative damage to the protein, which can occur when Fe3+metal ions are present in solution.Table 15: TEV-53408 Amino Acid (AA) Modifications Determined by Peptide Mapping Analysis

[0204] The risk of oxidation of W103, W109, M357 and M427 were statistically analyzed to predict an optimal formulation composition to decrease TEV-53408 oxidation is 0.1 mg / mL of PS80 and 0.2 mM of EDTA. The O. lmg / mL concentration of PS80, however, is too low to ensure interfacial stability of TEV-53408.

[0205] Table 16 summarizes specific quality attributes for optimal concentrations of PS80 and EDTA. The practical significance of PS80 and EDTA was evaluated based on data variability and the practical importance of the change in a measured quality attribute.

[0206] The PS80 concentration of 0.5 mg / mL was selected for further development, since no significant improvement of TEV-53408 quality attributes was observed with a PS80 concentration above 0.5 mg / mL, and that oxidation increased with increasing PS80 concentration. As a result, the combination of 0.5 mg / mL of PS80 with 0.1 mM of EDTA was selected for further formulation development.Table 16: Summary of Optimal Formulation Conditions Based On Different Quality Attributes of TEV-53408Note: ND = not defined due to the lack of statistical significanceExample 3: Turbidity of 100 mg / mL Formulation

[0207] The purpose of this study was to mitigate high solution turbidity which often signals risk of aggregation, and particle formation. Earlier experiments showed TEV-53408 was sensitive to ionic excipients, notably NaCl and, to a lesser extent, Arginine. Therefore, NaCl was eliminated, and the arginine concentration was reduced from previously established target of 75 mM down to 0 mM - 30 mM. In this experiment, Arginine-glutamate was used because it is less ionized than other arginine salts and because arginine alone did not previously show much benefit. Small changes in histidine and sucrose concentration and solution pH were also introduced to identify the impact on solution turbidity. A full factorial design was used (Table 17). Statistical data analysis was carried out using JMP 13 (version 13.2.1) software. The solution pH at initial preparation was within the target pH ± 0.15 for all formulations.Table 17: pH and Formulations Tested and Visual Inspection, Turbidity, and Colloidal Stability by DLS ResultsNote: ArgGlu = arginine-glutamate; O = opalescent; SO = slightly opalescent; VSO = very slightly opalescent

[0208] All samples were clear, slightly opalescent to opalescent solutions (Table 17). The detection of visible particles was complicated by formation of microbubbles upon swirling a vial.(results not shown). The formulations C01 and C03 appeared the least opalescent, had lowest turbidity, and lowest hydrodynamic radius compared to the rest of the formulations. This correlation suggest that high turbidity in TEV-53408 solutions can be due to strong protein-protein interactions.

[0209] A statistical analysis of the turbidity profile indicated that no arginine should be included in order to keep turbidity at an acceptable value of around 20 NTU units at 100 mg / mL protein concentration which, and that the optimal concentration is identified 20 mM histidine, 250 mM sucrose at pH 5.2.

[0210] As a result of the studies described herein and in Example 2, the formulation of 100 mg / mL TEV-53408 in 20 mM histidine with 250 mM sucrose, 0.1 mM EDTA, and 0.5 mg / mL polysorbate 80 at pH 5.2 was used for further formulation development.Example 4: Surfactant Screening for TEV-53408 Formulation Development

[0211] The purpose of this study was to compare Polysorbate 20 (PS20) and Polysorbate 80 (PS80) formulations for improved TEV-53408 compatibility and colloidal stability. The basic formulation for TEV-53408 was selected at target 100 mg / mL protein concentration in 20 mM Histidine, 250 mM Sucrose, 0.1 mM EDTA and 0.5 mg / mL of Polysorbate 20 or 80 at pH 5.2. Measured protein concentration in all formulations met the set target of 100 mg / mL ± 10%. Measured pH values stayed in the pH ± 0.1 range, and Measured Osmolality was 381 mOsmol / kg. The protein concentration, osmolality and pH measured in all formulations at TO were in close agreement with expected value.

[0212] Formulation stability was evaluated by subjecting samples to agitation (500 rpm for 24h at RT) stirring (150 rpm for 4h at RT) and 3 freeze-thaw cycles (Freezing at -20°C for > 8h; thawing at RT for > 8h.Appearance by Visual Inspection, Turbidity, and Colloidal Stability by PLS

[0213] All TEV-53408 formulation samples tested had many bubbles and it was difficult to distinguish visible particles, but foreign fibers were detected in all vials. Initial samples were clear, slightly opalescent solutions, with few fibers. Proper detection of visible particles was obscured by formation of micro bubbles after swirling the vials. The D02 formulation with PS20 had clear visible particles.

[0214] After agitation, the samples were clear, slightly opalescent solutions, with possible visible particles or fibers. After stirring, all samples were opalescent solutions, with visible particles or fibers. After 3 FT cycles, the samples were clear, slightly opalescent solutions.

[0215] Turbidity measurements for both formulations under all stress fell between 21 NTU and 23 NTU, except for samples after the stirring stress test. The formulation with 0.5 mg / mL PS80 had a turbidity of 25.4 NTU and with 0.5 mg / mL PS20 had a higher turbidity of 31.5 NTU.

[0216] As shown in Table 18, both formulations showed acceptable viscosity of around 5.8 cP, at 20°C temperature. No significant increase in the hydrodynamic radius was observed after the agitation or 3 FT cycles stress conditions compared to the initial time point. The stirring samples had an apparent increase in Rh by 0.8 nm for D01 and by 2.6 nm for D02. Polydispersity increased more than three times for the PS20, indicating broader distribution in hydrodynamic radius of the nano-particle species. The increase in polydispersity and turbidity indicates the formation of small particles in the range of 1-1000 nm and is associated with the formation of nano-aggregates with Polysorbate 20.Table 18: Viscocity, Hydrodynamic Radius (Rh) and Polydispersity (%Pd) Measured at Original Protein Concentration Under Different Stress Conditions

[0217] Protein-protein interactions (kD) measured by DLS in diluted samples in the concentration range of 4 - 24 mg / mL showed negligible differences between the formulations (PS80 was 0.53 mL / g, PS20 was 0.80 mL / g) that can be attributed to experimental error. The kD parameter measurements indicate slight repulsive interactions between molecules for both formulations.Thermal Stability by PLS

[0218] A dynamic light scattering (DLS) plate reader was used to monitor TEV-53408 aggregation during temperature ramping. Hydrodynamic radius (Rh) and polydispersity (%PD) were used to track changes in protein size and distribution of populations during the temperature increase from 25°C to 75°C, measured in 2.5°C steps, as shown in Table 19 and Figures 4A-4B.Table 19: Thermal Aggregation Parameters of TEV-53408 by DLS

[0219] Both formulations showed a similar thermal profile, with two transitions and no detectable full protein aggregation up to 75°C. The hydrodynamic radius remained in close agreement to the initial time point values. Slight changes in Rh were detected at around 52°C. Polydispersity thermograms showed a steeper aggregation curve for formulation with PS20 with a 10°C lower aggregation temperature than in formulation with PS80. In this case, the %PD detected full aggregation of TEV-53408 at 65°C for DOI (PS80) and at 55°C for D02 (PS20).Thermal Stability by Differential Scanning Calorimetry (DSC)

[0220] Temperature-induced unfolding of TEV-53408 was measured by differential scanning calorimetry (DSC) to assess protein structure and thermal stability between the two formulations containing PS80 andPS20. TEV-53408 at pH 5.2, 20 mM Histidine, 250 mM sucrose, 0.1 mM EDTA and 0.5 mg / mL PS80 or PS20 had Tonset (when the molecule begins to melt) is around 46.5°C and presented three separate transitions, with melting points (Tm) of approximately 52°C, 66°C, and 78°C. Only slight differences in enthalpy (the area under the peak) could be detected for the third transition (Fab).

[0221] These results suggest that the first unfolding event in the intact TEV-53408 is associated with the melting of the CH2 domain in the Fc fragment, with the second transition representing mainly the unfolding of the CH3 domain, and the melting of the Fab fragment showing the highest thermal stability in the third transition. These observations support the conclusion that the unfolding of the Fab and Fc fragments are independent events, and that the temperature transitions could be assigned to certain domains of the intact antibody, unless a detailed analysis of separate domains is made. Under closer inspection of Fab domain by deconvolution, the mathematical model reveals two transitions under one peak (see, Figures 5A- 5B).

[0222] Usually there is an assumption that the Fab fragment unfolds in a cooperative manner, that is, only one transition is observed in the thermogram of the Fab fragment. If the coupling among the domains in the Fab is disrupted by the CDR grafting and humanization process, the thermogram of the Fab fragment can present multiple transitions, and the interpretationof the DSC profile for an intact antibody will become more complex. Caution should be exerted when interpreting irreversible melting fragments. It could be possible that a correct evaluation of the area for the transitions that occur at high temperatures can be precluded by different contributions of the aggregation process to the thermogram.

[0223] Since there was no full aggregation detected in DLS hydrodynamic radius analysis up to 72.5°C, the reversibility of the TEV-53408 first two domains was tested by DSC (Figure 6). The temperature was raised to 58°C for the first domain, and then the cell was cooled and rescanned to 58°C again (dark lines). The first domain shows full reversibility. Slight loss in the area can be explained by short cooling time of the cell that did not give enough equilibration and refolding. For the second domain the temperature was raised to 68.5°C, the cell was cooled and rescan was made up to 68.5°C (light lines). Both domains showed little reversibility after the scan to 68.5°C. This indicates that the protein can withstand temperatures up to 58°C without completely aggregating and with reversible folding, but the prolonged exposure to high temperatures can still induce higher aggregation rates in the TEV-53408.

[0224] Figure 7 shows the full thermal stability profile for TEV-53408 in PS80. From the DLS and DSC data it was observed that upon the beginning of second domain melting, an increase in hydrodynamic radius (Rh) is detected that denotes denaturation and aggregation of the molecule.Sub-Visible Particulate Matter by Micro-Flow Imaging (MFI)

[0225] MFI data for the TEV-53408 formulation samples for TO and after stress did not show a clear effect of PS80 vs PS20 in the TO versus Agitation and Freeze-thaw samples, with variations attributable to method variability. In contrast, stirring samples showed significant differences in the formation of particles <50 pm. Following stirring, the formulation with PS80 showed insoluble aggregates in 1 - 25 pm size range, whereas the formulation with PS20 showed formation of smaller particles within the range of 1-1000 nm with increased turbidity of the solution and visible particle formation that were detected during visual inspection.Purity assessments.

[0226] Purity, as assessed by Size Exclusion Chromatography (SE-HPLC), Non-Reducing Capillary Gel Electrophoresis (CGE), or Capillary Iso-electric Focusing (cIEF), showed no differences between PS80 and PS20 samples, and no changes between initial and stressed condition.

[0227] Representative data (from the PS80 formulation):(a) SE-HPLC showed 98.75% monomer, 1.16% dimer, no detectable higher order aggregates; and 0.09% low molecular weight species;(b) non-reduced CGE showed 99.2% IgG, 0.3% HHL (fragmented heavy and light chain), and 0.5% fragments. High molecular weight species were not detected; and(c) charge heterogeneity by cIEF showed 23.3% Acidic Peak, 72.8% Main Peak, 3.9% Basic Peak.

[0228] In conclusion, PS80 and PS20 surfactants performed similarly in most tests, but PS80 showed superior performance in DLS, turbidity and visible particle formation. Accordingly a final formulation selected with Polysorbate 80 was as follows: 100 mg / mL TEV-53408 in 20 mM Histidine buffer, 250 mM Sucrose, 0.1 mM EDTA, 0.5 mg / mL Polysorbate 80 at pH 5.2.Example 5: In-Use Compatibility and Stability of TEV-53408 Drug Product

[0229] The clinical in-use compatibility and stability of TEV-53408 drug product (100 mg / mL TEV-53408 in 20 mM Histidine buffer, 250 mM Sucrose, 0.1 mM EDTA, 0.5 mg / mL Polysorbate 80 at pH 5.2.) was tested for subcutaneous (SC) and intravenous (IV) routes of administration.Subcutaneous

[0230] To model SC injection, TEV-53408 drug product was withdrawn from upright vials using 18G needles into ImL and the 3mL sterile disposable syringes of polycarbonate (PC) or polypropylene (PP), and capped with 27G needles syringes. After under normal laboratory light conditions at 15-25°C for, the antibody was stable up to 6 hours. The control sample for the SC arm was the TEV-53408 drug product in the 6R vial stored at 2-8°C. In addition, vials of TEV- 53408 were subjected to one freeze-thaw cycle to evaluate the impact of accidental freezing on the drug product quality. The freeze-thaw arm was prepared by placing vials into a -20°C freezer for 24 hours and thawing at room temperature (15- 25°C) for at least 2 hours.

[0231] All samples were essentially free of visible particulates, and were equivalent in opalescense, protein concentration, pH and osmolality (Table 20). Other tests showed no significant changes were caused by storage. No significant changes were observed between samples when examined for sub-visible particulate matter by light obscuration (HIAC) and microflow imaging (MFI); for the distribution of species measured by SEC, reduced or non-reduced CGE; charge heterogeneity by cIEF; or potency by NK cell based assay (results not shown).

[0232] The data indicate that in-use storage and one freeze-thaw cycle does not have any significant impact on the properties of the TEV-53408 solution in syringes.Table 20: TEV-53408 Sample Description SC In-Use Compatibility StudyIntravenous

[0233] In-use stability and compatibility for IV administration was studied using TEV- 53408 diluted to between 1 to 20 mg / mL in common diluents 0.9% w / v sodium chloride (normal saline), or 5% dextrose solutions, in IV containers made from polyvinyl chloride (PVC) or polyolefin (PO), Solutions were collected through the IV administration set from IV bags immediately after dilution (TO time point). The IV bags with diluted TEV-53408 DP were stored in a biosafety cabinet in the hanging position exposed to artificial light at 15-25°C for 8 hours (only for 5% dextrose as a diluent) and for 24 hours (both dextrose and normal saline). Samples were collected, via gravity feed, from the bags within the 2 hour incubation time. A control was created by completing the same process as was done for the sample but using diluent instead of TEV- 53408 DP.

[0234] As shown in Table 21, TEV-53408 drug product was compatible upon dilution with 5% dextrose, and was stable up to 24 hours when stored at 15-25°C under ambient light, in both PVC and PO bags. No particles were observed on visual inspection and the turbidity remained low. This shows acceptable in-use stability profile.

[0235] In contrast, TEV-53408 drug product was not compatible with 0.9% w / v sodium chloride (normal saline) as a diluent because of formation of visible particles (not shown) The formation of visible particles corresponds to the increased relative turbidity. The presence ofvisible particulate matter after TEV-53408 dilution with 0.9% sodium chloride is a patient safety factor, thus the TEV-53408 drug product cannot be diluted with 0.9% sodium chloride.Table 21: TEV-53408 Sample Description for the In-Use Compatibility Study of IV formulationsNotes: RT = room temperature (15-25°C); PVC = polyvinyl chloride; PO = polyolefin; IV = intravenous; Low Cone. = nominal Img / mL TEV-53408 concentration; High Cone. = nominal 20mg / mL TEV-53408 concentration

[0236] Dilution in saline was also associated with increased sub-visible particulate as measured by MFI and HIAC, in terms of the concentration of sub-visible particles with size >2pm and >10pm based on the MFI analysis and for particles with size >2pm, >10pm and >25pm based on the HIAC analysis. Dilution in dextrose did not increase particles as measured by MFI and HIAC. No clear trends are observed for sub-visible particles as a function of storage time. No changes after dilution or storage was observed for: purity, as measured by SEC, and CGE (reduced and non-reduced); or charge heterogeneity measured by icIEF. These show that the antibody is stable in the diluents.

[0237] The relative potencies for the samples was within specifications, whether prepared in either 0.9% sodium chloride or 5% dextrose as compared to the initial (TO) sample. Although samples diluted to 1 mg / mL with 5% dextrose showed relative potencies remained in the acceptable range, the values tended to be lower, suggesting that samples with more dilute protein concentrations (<1 mg / mL) need to be examined for potency.Conclusions

[0238] The TEV-53408 drug product in-use storage conditions for subcutaneous administration showed no significant change in solution pH, osmolality, protein concentration, purity by SEC and CGE (reduced and non-reduced), charge heterogeneity by icIEF, and cell based potency. These results support the conclusion that TEV-53408 DP is compatible with disposable polypropylene and polycarbonate syringes equipped with 27G needles. The TEV-53408 drug product can be stored in disposable syringes up to 6 hours when stored at 15-25°C and ambient light.

[0239] Upon dilution of the TEV-53408 drug product with 0.9% sodium chloride in PO or PVC bags, visible particles were observed at the initial time point (TO) and after 24-hour storage. In addition, an increase in the concentration of sub-visible particles and solution turbidity was detected. These findings suggest that the TEV-53408 drug product is not compatible with 0.9% sodium chloride diluent solution.

[0240] Upon dilution of the TEV-53408 drug product with 5% dextrose in PO or PVC bags, no significant changes were detected in this study, suggesting that TEV-53408 drug product is compatible with 5% dextrose diluent solution up to 24 hours after dilution when stored at 15- 25°C and ambient light. Therefore, the TEV-53408 drug product is compatible with a 5% dextrose diluent solution.Example 6: Stability of TEV-53408 Drug Product in 6R Vial

[0241] The purpose of this study was to evaluate the stability and compatibility of TEV- 53408 nominal liquid formulation in a 6R vial drug product configuration upon storage at the recommended temperature of 5±3°C, and at the stressed storage conditions of 25±2°C / 60±5% RH, and 40±2°C / 75±5% RH. At the predetermined time intervals, samples were analyzed for sub- visible particle count by micro-flow imaging (MFI) and biophysical characterization. Biophysical characterization of TEV-53408 was also carried out using DLS, rheometry, and circular dichroism (CD) measurements.

[0242] After manufacturing, the TEV-53408 drug product was stored at 5±3°C, protected from light, until stability study initiation. Glass vials were loaded in stability chambers in an inverted orientation.Dynamic Light Scattering (DLS) and Viscocity

[0243] DLS was used to assess potential changes in the hydrodynamic radius of active protein and monitored particle formation in nano-size range in solution. No significant changes beyond normal experimental variation were identified in hydrodynamic radius or polydispersity for samples stored at 2-8°C up to the specified storage times of 24 months (Table 22). There is a slight increase following storage at 25°C. Storage at 40°C was associated with a significant increase in Rh and %Pd over time. The main population (monomeric) hydrodynamic radius changed during stressed conditions and additional populations at higher Rh were clearly resolved after 3 -months of storage and can be attributed to higher order aggregates.Table 22: Viscosity and DLS Parameters of TEV-53408 Stability SamplesNote: NT = not tested

[0244] To assess viscoelastic properties of the molecule and possible aggregation during storage, viscosity measurements were collected using a Vi scorn eter-Rheometer-on-a-Chip (VROC) - Initium instrument (RheoSense, San Ramon, CA). The initial sample at 107 mg / mL protein concentration had a viscosity of 6.07cP when measured at 20°C. Storage at 2-8°C did not significantly increase viscosity. Increasing viscosity trends were observed at all stressed conditions (25°C and 40 °C), indicating potential aggregation due to protein-protein interactions. Regression analysis (not shown) showed a line of fit with 95% confidence intervals for the time of storage.Viscosity as a function of concentration and temperature.

[0245] Concentrated protein solutions (greater than 150 mg / mL) were prepared to characterize viscoelastic properties with changes in TEV-53408 concentration and temperature. The resulting protein concentrations were determined using UV-vis spectrophotometry absorbance at 280 nm, and viscosity measured using the VROC Initium as a function of protein concentration (Table 23) and temperature (Table 24).Table 23: TEV-53408 Protein Concentration and Viscosity MeasurementsTable 24: TEV-53408 DP Viscosity Measurements at Different Temperatures and Derived Parameters for Arrhenius Fit

[0246] The concentration-dependent increase in viscosity fit the semi-empirical Ross- Minton equation, suggesting no forces, other crowding / excluded volume, significantly affect viscosity in the TEV-53408 DP. The viscosity versus temperature data for the TEV-53408 DP solution fits the Arrhenius-type model. Viscosity rose from about 4.6 to 18.8 cP as the temperature was reduced from 25°C to 5°C.Sub-Visible Particle Determination by Micro-Flow Imaging (MFI)

[0247] The particle concentrations remai'ned at low levels and suggest no significant or unacceptable increase in particle count after 24 months storage at 2-8°C, 25°C, or 40°C. The number of sub-visible particles / mL remained within USP <788> acceptance criteria for particles >10 pm and >25 pm and remain within these criteria even if recalculated for the specified vial volume (particles per container) of 3.0 mL.Secondary and Tertiary Structure Assessment by CD

[0248] CD measurements are extremely effective for structural analysis and stability evaluation of proteins. CD spectra were measured with a JASCO spectropolarimeter model J- 1500 (JASCO International Co., Tokyo, Japan) in triplicate at far UV and near UV. The CD spectra were similar to that of a typical immunoglobulin, with a negative band at 215 nm and a zero intensity at a wavelength of 208 nm, representing a high content of P-sheet and several smaller positive bands at 201nm and 192 nm (not shown). For the protein solutions with weak far-UV CD signals (e.g., P-sheet protein like IgG) and / or buffer excipients that had absorbance or CD signals of their own in the far-UV region, especially <190 nm, the average spectral similarity is usually low and can vary greatly.

[0249] To estimate the content of the various structural elements, a fitting procedure was applied and the structural composition of the TEV-53408 lyophilized DP is presented in Table 25. TEV-53408 was found to consist of 46.3 % P-sheets, 20.6 % turns, 30.5 % unordered and 2.5% a- helical structures as calculated by software.Table 25: Secondary Structures for TEV-53408 Initial SamplesNotes: AVG = average; SD = standard deviation; Cv = coefficient of variation

[0250] The structural changes reflected in the near-UV region of the CD spectra are usually associated with reorientation of the aromatic amino acids tyrosine and tryptophan, and from the asymmetric environment of disulphide linkages. All three replicate measurements were similar, suggesting that the near-UV CD spectra obtained with the same cuvette are sufficiently precise for biopharmaceutical protein characterization. The most variability in the near UV CD spectra is for the same sample measured on the same day.

[0251] In the near-UV CD region negative absorption bands were observed between 255 to 270 nm, corresponding to phenylalanine (Phe), between 274 to 282 nm the absorption bands corresponded to tyrosine (Tyr), maximum absorption corresponding to the Tryptophan (Trp) residues observed between 285 to 305 nm.

[0252] Stability sample spectra of 3, 6, 18, and 24 months at different storage conditions were overlaid with the 95% confidence interval of 3 replicate scans of initial sample, Savitzky- Golay Linear smoother and Standard Normal Variate were applied to all spectra. Spectra before and after processing with 95% confidence interval are shown Figures 8A-8B. Most spectra fell within the 95% CI after smoothing. There is a significant shift from CI for the 6 months stability sample at 40°C where changes in higher order structure were expected, and this shift can suggest a significant change in the tertiary structure of TEV-53408.

[0253] These shifts were not associated with any secondary structure perturbations since none of the secondary structure elements changed significantly over time, as summarized in Table 26Table 26: Secondary Structures for TEV-53408 Stability SamplesThermal Stability by Differential Scanning Calorimetry (DSC)

[0254] DSC scans were performed during stability testing of the TEV-53408 liquid formulation to detect any conformational changes appearing during long-term storage. Two full scans at 1.0 mg / mL protein concentration for each sample were analyzed. TEV-53408 at pH 5.2,20 mM Histidine, 250 mM sucrose, 0.1 mM EDTA and 0.5 mg / mL PS80 presents three separate transitions with melting points (Tm) around 53.0°C, 66.6°C and 79.2°C. Tonset of the molecule (start of denaturation / melting) is around 47°C. The freshly prepared liquid protein formulation is 99.1% native (folded) protein by SEC. Slight differences in thermograms can be due to experimental variability.

[0255] As the protein sample deteriorates during storage, the fraction of native protein in the solution begins to decrease, resulting in a decrease of enthalpy in the DSC thermograms. DSC thermograms were taken at TO, 3, 6, 12, 18, and 24 months’ time points to assess any structural reorganizations in the molecule. Figure 9 shows an overlay of all data grouped by storage temperature before and after processing of the thermograms. Savitzky-Golay Linear smoother and Standard Normal Variate were applied to the thermograms to eliminate additive and multiplicative effects. After processing, the thermograms were normalized and more comparable. All scans at 5°C storage overlay well with initial sample up to 18 months and slight shift detected at 24 months storage in all domains. More data points were needed to confirm changes at 24 months stability point. Samples at 25°C storage started to shift to the first domain at 6 months, where purity by SEC is 98.5%. The most significant shifts were observed at the 40°C storage condition, as expected of TEV-53408, which has a low Tonset ~ 47°C. Samples stored at 40°C have much lower enthalpy for domain 1 and some shifts in domain 2 that indicates the loss of native structure and disruption of native bonds. This was expected at 40°C storage where purity by SEC is 95.6% at 3 months and 92.7% at 6 months. DSC is more sensitive than CD to changes in native structure and can detect much lower impurity levels, for example, impurity levels < 5%.

[0256] Stability sample thermograms at 3, 6, 12, 18 and 24 months at the different storage conditions overlaid with the 95% confidence interval (low variability of DSC measurements) constructed from two scans of the initial samples are shown in Figure 10. Arrows indicate changes in enthalpy during stability.

[0257] DSC is more sensitive technique to structural changes and especially to aggregation than CD, and can detect changes at 2% aggregate levels. Table 27 summarizes the parameter changes observed during the stability testing. The main parameters extracted from the DSC thermograms were Tonset and melting temperatures of each domain (Tm). There was a measured dependency between the parameters and the change in storage temperatures and time, with a reduction in Tonset and Tm3(Fab) over time and due to higher storage temperatures, suggesting degradation of the molecules over time. In addition, Tm2 shifted to higher temperatures at 40°C storage, which was also observed in the thermograms.Table 27: DSC parameters for stability samples using option ‘Pick Peaks’ Stability pointConclusion

[0258] All quality attributes showed no significant changes when the liquid TEV-53408 drug product was stored at 2-8°C for up to 24 months. Temperature and time dependent chemical degradation and / or aggregation was observed by DLS and viscosity measurements. Most changes were detected at the 40°C storage condition, with significant formation of aggregates observed as increases in polydispersity and hydrodynamic radius. Increases in viscosity also suggested the formation of protein-protein interactions. The concentration dependence of solution viscosity was used to determine the apparent intrinsic viscosity, shape, and “crowding” factors for TEV-53408 DP. The temperature dependence of the solution viscosity was quantified further in terms of the effective activation energy of viscous flow that indicated microstructure formation. MFI indicated that there was no temperature or time dependent trend in sub-visible particle counts and levels stayed within acceptable ranges.

[0259] Biophysical characterization by DSC and CD did not show any significant structural perturbations during 24-month storage at 5°C. DSC and CD measurements can typically be correlated with aggregate formation in the sample, and SEC can typically be used as a reference method to link any structural changes. DSC can typically detect structural changes when there is <2% of aggregation, denaturation or truncation appearing in the molecule. Samples stored at 25°C and 40°C exceeded the 2% threshold for higher order aggregates by SEC, or by CGE, indicating that changes in the thermograms in the first domain could be attributed to the degradation of TEV-53408. DSC parameters also changed during stability tested, which could indicate degradation of the molecule. The CD method has a higher threshold for structural sensitivity (~5%), and therefore only changes at the 40 °C storage condition were detected in the tertiary structure by near UV measurements.

[0260] Overall, the data was well within the evaluation criteria that indicates that the nominal liquid TEV-53408 drug product is stable and compatible with a Type-I glass 6R vial with a rubber stopper for up to 24 months when stored at 2-8°C. TEV-53408 has a low Tonset compared to many other antibodies, perhaps due use of IgG4 with YTE modifications. Hence, the molecule is sensitive to higher temperatures, and degradation was observed after 3 months of storage at 40°C, with some stability changes also observed after extended storage at 25°C.Example 7: High Concentration TEV-53408 Drug Product Formulation

[0261] The purpose of this study was to evaluate the stability and compatibility of several viscosity reducing excipients added to the base formulation of TEV-53408 at 100 and > 150 mg / mL in 20 mM histidine, 250 mM sucrose, 0.1 mM EDTA and 0.05% PS80 at pH 5.2.

[0262] Eleven excipients were selected for the viscosity lowering study (Table 28). The impact of these excipients was evaluated using TEV-53408 DP.Table 28: Excipients and Concentrations Evaluated in Viscocity Lowering StudyND: Not determined

[0263] Viscosity for 105 ± 3 mg / mL TEV-53408 was measured at 20°C. Excipient viscosity lowering effects were compared to the initial DP formulation including sucrose by performing statistical analysis (Dunnett’s test) and comparing LSMeans to Control sample - DP. Statistical analysis showed significant changes in viscosity for almost all excipients except glycine and lysine. NaCl gave a significant rise in viscosity. This excipient was excluded from further evaluation for this reason, and because previous experiments showed a tendency of NaCl to cause aggregation of TEV-53408. Excipients that showed very small decreases in viscosity. Arginine, Leucine, Proline, MgCl and Histidine were selected for further development. To confirm the viscosity lowering effects over the range of protein concentrations, formulations with the best performing excipients were concentrated using Amicon Ultra- 15 centrifugation units, and the viscosity curve vs protein concentration was constructed. All formulations performed similarly up to 120-130 mg / mL. At 150 mg / mL all formulations passed 30cP (Figure 11), which is often considered the practical limit for handling drug product. Following overconcentration of the bulk drug substance (BDS) up to 180 mg / ml, the excipients that showed viscosities lower than 70cP were histidine, MgCL and, most significantly, arginine. Viscosity reduction was also dependent on arginine concentration: at 180 mg / ml protein concentration the viscosity of the solution was approximately 53cP for 130mM of arginine, and approximately 45cP for 200mM of arginine (Figure 11). Further development with arginine hydrochloride as a viscosity reducing excipient was explored to establish an optimal high concentration TEV-53408 formulation.

[0264] Concentrated TEV-53408 DS (192.2 mg / mL protein concentration, 0.1% PS80, 140 mM arginine hydrochloride, and pH 52, having a measured viscocity of 52.7 cP) was prepared by ultrafiltration / diafiltration with an exchange buffer consisting of 20 mM histidine, 175 mM arginine hydrochloride at pH 5.2, and used to prepare the formulations in Table 29. The measured protein concentration in all formulations was within the target ± 3 mg / mL and the measured pH values were within the 5.2 ± 0.1 range (not shown).

[0265] Stress conditions tested with the high concentration TEV-53408 formulations consisted of: i) 25°C±2°C / 60±5% RH for 3 months; ii) 40°C±2°C / 75±5% RH for 1 month; and iii) freezing at -20°C for > 24h; thawing at RT for > 24h (5 cycles). Two 6R vials per formulation were prepared for each of the stress tests. Appearance results by visual inspection are summarized in Table 29.Table 29: High Concentration Formulations and stability

[0266] Initial samples were opalescent solutions, essentially free from visible particles, some vials contained foreign fibers. After 1 month at 40°C storage all of the samples were highly or very opalescent solutions. After 3 months at 25°C storage all of the samples were opalescent or very opalescent solutions. 5 FT cycles did not change the appearance.

[0267] Turbidity was evaluated with a HACH turbidimeter, using a calibration curve with three NTU standards: < 0.1, 20 and 200 (Table 29). All formulations were more opalescent than 20 NTU, with the lowest values measured for formulations E04 and E10, which contain the highest arginine and sucrose concentrations. The turbidity did not increase significantly after 5FT and storage for 3 months at 25°C as compared to the initial time point. 1 month storage at 40°C increased turbidity.

[0268] Sucrose was found to be the most significant variable that works towards lowering turbidity in the initial samples and after stress conditions. Protein concentration and arginine have a slight turbidity reducing effects.Viscosity

[0269] The viscosity of the high concentration TEV-53408 formulations is summarized in Table 30. As a general rule, injectable drugs usually target viscosities below 25 cP. All formulations at 150 mg / mL (E01-E04) had acceptable viscosities while those with 170 mg / mL(E07-E10) had viscosities in the range of 21 - 27 cP. Viscosity was greatly increased by cooling to a standard storage temperature of 5 °C, indicating that calibration to at least room temperature would be required before administration.Table 30: Viscocity of High Concentration TEV-53408 Formulations at Different Storage Conditions

[0270] Storage for 3 months at 25°C did not have any significant effects on viscosity, indicating a stable TEV-53408 composition with no changes due to aggregation or additional protein-protein interaction formation. One month of storage at 40°C increased viscosities, indicating changes in the protein-protein network during stability, most likely due to soluble aggregate formation. The change (Delta=lM@40°C-Initial) in viscosity was calculated to evaluate the stability of the formulations. Protein concentration was the main variable that highly increased viscosity of the solution. Sucrose had a significant viscosity reducing effect at 40°C storage that can be associated with reduced aggregate formation, and support a role for sucrose as stabilizer under elevated temperatures. Arginine did not play a very significant role in the explored range of 150 - 200 mM.Thermal Stability by PLS and DSC

[0271] TEV-53408 aggregation was monitored by DLS during temperature ramping from25°C to 72.5°C in 2.5°C increments. The hydrodynamic radius (Rh) and polydispersity (%Pd) data is shown in Table 31 and the thermograms in Figures 12A-12B. The hydrodynamic radius as afunction of temperature shows two thermal transitions (Figure 12A). The thermal stability data show some differences between TEV-53408 formulations with E03 and E09 being least stable and E02 most stable.Table 31: Aggregation Onsets (Tonset) and Full Aggregation (Tagg) Temperatures by DLS and Tonset and Tm by DSC

[0272] DSC analysis was performed only on the first six formulations, since the method requires dilution of sample to 1 mg / mL of protein concentration and this eliminates the effect of protein concentration. Statistical analysis predicted that protein concentration, protein concentration* sucrose (that is, the combination of concentrations of protein and sucrose), and arginine were found to be the most significant for thermal stability. Sucrose is the main variable that highly stabilizes the molecule.

[0273] The DLS analysis results for TEV-53408 in different formulations are summarized in Table 32. To assess protein-protein interactions, the kD interaction coefficient was measured by DLS in the concentration range of 3 - 32 mg / mL. kD was only calculated for formulations F01- E06, since only these formulations have differences in excipients. The kD for formulation E01 with just 150mM arginine was -9.45 mL / g; for other formulations the kD was around 7 mL / g within method variability. The kD variable indicated attractive interactions between molecules for all formulation. Also the addition of arginine increased hydrodynamic radius of the molecule from 7 nm to ~13 nm. Statistical analysis indicated that arginine, arginine* sucrose interaction and sucrose were found to be the most significant variables for TEV-53408 colloidal stability. Sucrose is the main variable that highly stabilizes the molecule at elevated temperatures. The analysis furthersuggests that the formulation with best thermal stability which can be achieved with 150 mg / mL protein concentration, 200 mM arginine HC1 and 200 mM sucrose. All formulations at 40°C were found to be multimodal systems (Pd>50%) with formed aggregates at 90-100 nm. Other stability conditions did not indicate significant changes in Rh and %Pd.Table 32: Aggregation and protein-protein interactions by DLSNotes: NT = not tested; TO = initial time pointSub-Visible Particles by Micro-Flow Imaging (MFI)

[0274] The MFI data for TEV-53408 formulations at the initial and stressed conditions are summarized in Table 33 and Table 34. The sub-visible particle concentration is similar in all formulations at initial time point and after stressed conditions. Furthermore, no significant differences were observed among all formulations for sub-visible particles at all conditions by Dunnett’s test (data not shown).Table 33: Concentration of sub-visible particles by MFI in high concentration TEV-53408 formulations at initial time pointTable 34: Concentration of sub-visible particles by MFI in TEV-53408 formulations after IM at 40°C and 5 freeze-thaw cycles (5FT)Purity by Size Exclusion Chromatography (SE-HPLC) and CGE

[0275] TEV-53408 purity analysis by SE-HPLC are summarized in Table 35. At initial time point the % monomer was 99.2%, the % dimer was 0.7%, % LMW was 0.0.6-0.11%, with no detectable HOA across all formulations, before or after stress. Purity by SE-HPLC was not significantly affected by 5 freeze thaw cycles (not shown) but was impacted by storage for 1 month at 40°C and 3 months at 25°C (Table 35). The highest degradation was observed at 40°C: the % monomer decreased by approximately 3-4%, the % dimer increased by approximately 3-4% and the fragments increased up to 0.3-0.4% with no change in % HOA.Table 35: Purity of TEV-53408 by SE-HPLC at initial time and after stressNotes: TO = initial time point; LMW = low molecular weight species

[0276] Sucrose, arginine, protein and protein* sucrose interaction were determined to be the most significant variables in a statistical model for purity by SEC. Sucrose acts as stabilizer and prevents aggregate formation in solution. The statistical analysis suggests the best formulation for highest stability by SE-HPLC is at 150 mg / mL protein concentration, including 150 mM arginine HC1 and 200 mM sucrose.

[0277] The results of TEV-53408 purity by CGE under non-reducing conditions are summarized in Table 36, and under reduced conditions in Table 37. At the initial time point the purity of TEV-53408 by non-reducing capillary gel electrophoresis (CGE) was 99.3% of IgG, 0.4% of fragments, and 0.4% of HHL truncated species in all formulations. For reducing CGE, TEV- 53408 at the initial time point had 98.2-98.6% LC+HC and 0.6-0.7% NGHC. No significant changes were observed in any of the formulations after 5 FT stress (not shown) (Table 37). Formulations after storage at 40°C and 25°C showed degradation mostly by forming fragments as detected by non-reduced CGE. The highest change in purity in all formulations at different stress conditions was observed at 40°C storage at both non-reduced and reduced conditions. Formulations at 25°C storage showed changes in purity only under non-reducing conditions.Table 36: Purity of TEV-53408 by Non-Reducing CGE for TO and after stress conditionsNotes: HOA = higher order aggregates; TO = initial time pointTable 37: Purity of TEV-53408 by Reducing CGE for TO and After 1 Month at 40°C and 3Months at 25°CNotes: TO = initial time point; LC = light chain; HC = heavy chain; NGHC = non-glycosylated heavy chain

[0278] The statistical data analysis results with reduced model and best fit (not shown) indicated that sucrose concentration is the main variable that highly effects the stability of TEV- 53408. Arginine on the other hand has slightly different effect on protein stability depending on storage conditions, but it was the main variable to increase fragmentation of TEV-53408 at elevated temperature (40°C) storage.Charge Heterogeneity by Imaged Capillary Iso-Electric Focusing (icIEF)

[0279] The charged species distribution in the high concentration TEV-53408 formulation samples is shown in Table 38 for TO and samples after 1 month at 40°C and 3 months at 25°C. The distribution of TEV-53408 charged species at TO is essentially the same in all formulations at TO, and no significant change was observed after the 5FT stress condition. After storage for 1 month at 40°C and 3 months at 25°C, the main isoform decreased in all of the formulations, indicating degradation of the TEV-53408 drug product.Table 38: Charge Heterogeneity in TEV-53408 Formulations in Initial Samples and stressNote: TO = initial time pointCell-Based Potency Assay

[0280] The relative potency for the TEV-53408 high concentration formulations is shown in Table 39. All relative potencies for the samples were within the acceptable. No apparent change was observed between formulations at the initial (TO) time point. After storage for 1 month at 40°C, the samples showed somewhat lower relative potency in the range of 87 - 101%. These results are within the acceptable range for the TEV-53408 drug product. Differences in the cell potency of the formulations can be related to increased degradation of the drug product under the stress condition that can lead to the lower potencies.Table 39: Cell Based Potency for the TEV-53408 High Concentration FormulationsNote: TO = initial time point; NT = not testedConclusions

[0281] The optimal concentrations of protein, sucrose and arginine in the TEV-53408 high concentration formulations based on specific quality attributes are summarized in Table 40. The practical significance of each variable was evaluated based on data variability and the practical importance of the change in a measured quality attribute by varying the statistical predictors in each experiment.Table 40: Summary of Optimal High Concentration Formulation Variables Based on Different Quality Attributes of TEV-53408Note: NA = not applicable

[0282] The analysis based on statistically and practically significant TEV-53408 quality attributes indicated the following conclusions can be drawn about the formulation variables toachieve overall TEV-53408 HCF stability and acceptable viscosity at 150 mg / mL and >150 mg / mL: the preferred Arginine HC1 concentration is 180 mM since higher Arginine concentrations are needed to reach proper viscosities at > 150 mg / mL (BDS process); the preferred Sucrose concentration is 200 mM; the suggested formulation should have a theoretical osmolality >600 mOsmol / kg. Accordingly, based on the global desirability profiler (data not shown) and optimal variables (Table 40) the formulation can be adjusted to include 180 mM Arginine HCL and 180 mM Sucrose. A suitable TEV-53408 high concentration formulation can therefore comprise 150 mg / mL TEV-53408, 20 Mm Histidine, 180 mM Arginine hydrochloride, 180 mM Sucrose, 0.1 mM EDTA, 0.5 mg / mL PS80, pH 5.2. The osmolality of such a target concentration formulation at 150 mg / mL is around 690 mOsmol / kg and viscosity is around 16 cP at 20°C, in line with the acceptable specifications for TEV-53408 high concentration formulation. Osmolality and viscocity measurements for TEV-53408 high concentration formulations prepared from a BDS consisting of 150 mg / mL protein concentration, 20 mM histidine, 180 mM arginine HC1, 180 mM sucrose, 0.1 mM EDTA and 0.9 mg / mL PS80, and pH 5.2 are shown in Table 41.Table 41: Osmolality and Viscocity for Target High Concentration FormulationsNote: FB=Formulation bufferExample 8: Development of In Serum TEV-53408 Detection Methodology

[0283] To detect TEV-53408 in serum over a wide range of concentrations, two separate detection methods were developed and validated.Preparation of calibration standards (STDs) and quality controls (PCs)

[0284] STD and QC were prepared by spiking TEV-53408 into 100% pooled normal human serum (PNHS). The standard curve in human serum ranged from 50 to 15000 ng / mL for detection method 1, and from 2 to 2000 ng / mL for detection method 2. Five concentration levels of QCs were prepared at 100 (LLOQ), 300 (LQC), 500 (MQC), 7500 (HQC), and 10000 ng / mL (ULOQ) for detection method 1, and at 5 (LLOQ), 15 (LQC), 150 (MQC), 1400 (HQC), 1800 ng / mL (ULOQ) for detection method 2 (LLOQ = lower limit of quantification; ULOQ = upper limit of quantification; LQC = low quality control; MQC = mid quality control; HQC = high quality control). The STDs were either prepared fresh on the day of the run or prepared in advance and stored in a -80°C freezer. QCs were prepared in advance and stored in a -80°C freezer to mimic the incurred samples.Method development

[0285] 5 anti-IgG4 monoclonal antibodies and 2 anti-human IgG polyclonal antibodies, at1 pg / mL, were screened to identify the best antibody pairs based on signal to noise ratio and background signals. The assays were further optimized for the concentration of capture and detection antibodies, the minimum required dilution (MRD), and blockers to ensure adequate sensitivity, selectivity, and interference. The wash buffer was prepared by diluting 20X Wash Buffer (Alpha Teknova Corp., CA, USA) 1 :20 with DI water. The optimal dilution buffer for detection method 1 is 1% Casein in PBS, 0.1% Tween-20 supplemented with 5% horse serum. The optimal dilution buffer for detection method 2 is 1% Casein in PBS, 0.1% Tween-20, and 200 pg / mL TRU Block 3 (Meridian Bioscience Inc., TN, USA). Method validation was consistent with guidelines from FDA, European Medicines Agency (EMA), and ICH M10 guidance for industry.TEV-53408 detection method 1

[0286] A 96 well Streptavidin Sector microplate was blocked with blocking buffer (1% Casein in phosphate buffered saline (PBS)) at room temperature for Ihr (hour), then coated with Ipg / mL of biotinylated anti-IgG4 antibody (AbD36430) for Ihr with shaking. STDs, QCs, and validation samples were diluted at 1 : 100 MRD with dilution buffer and added to wells, incubated for Ihr. The plate was washed and 0.5 pg / mL of the sulfo-tag labeled anti human IgG antibody was added to the plate. After Ihr incubation, the plate was washed, read buffer was added. Electrochemiluminescence is measured by a plate reader. The standard curve fitting model is the 4 PL Marquardt with 1 / mean value2(1 / Y2) weighting factor performed in Watson LIMS. Theconcentrations of MabOl were interpolated from the standard calibration curve. The mean of the concentrations of the duplicates was reported.TEV-53408 detection method 2

[0287] Method 2 had a similar experimental procedure as method 1 except for the dilution buffer. In addition, the capture reagent, biotinylated anti-ID antibody was used at 2.5 pg / mL, and the sulfo tag conjugated anti-human IgG antibody detection reagent concentration was 2 pg / mL. STDs, QCs, and samples were diluted at 1 :30 MRD with dilution buffer and let sit at least 15 minutes at 2-8°C before adding to the plate.Validation of PK method 1 and 2

[0288] The accuracy, precision and selectivity was confirmed for both methods, and is summarized in Table 42. For method 1, the standard curve range was validated from 100 to 10000 ng / mL, with 50 and 15000 ng / mL as anchor points, with a sensitivity of 100 ng / mL; and for method 2, the range of the calibration curve spanned from 5 to 1800 ng / mL, with 2 and 2000 ng / mL as anchor points, with a sensitivity of 5 ng / mL. Both methods showed specificity and a linear dose.Table 42: Precision and accuracy of method 1 and 2QC = quality control; %RE = percent relative error; %CV = percent coefficient of variation; n = number; LLOQ = lower limit of quantification; ULOQ = upper limit of quantification; LQC = low quality control; MQC = mid quality control; HQC = high quality control; |Mean| = absolute mean value; ng / mL = nanograms per milliliterFree and total human IL- 15 assays

[0289] Methods were developed to measure serum free IL-15 (i.e. drug-unbound IL-15 / IL- 15Ra) and total IL-15 (free IL-15 and IL-15 bound to TEV-53408). Free IL-15 in serum was quantified using a ligand-binding assay. On a Streptavidin-coated 96-well plate, a biotinylated antihuman IL-15 antibody that is competing with the TEV-53408 was used to capture only drug- unbound free IL-15 / IL-15Ra, which is the dominant form of IL-15 in human serum. Sulfo-tagged TEV-53408 was used as the detection antibody to complete immune complex formation. By using the human IL- 15 specific capture / detection antibody pair and a recombinant human IL-15 / IL- 15Ra calibrator, the assay measured drug unbound, free IL-15 / IL-15Ra. Qualification results for the free IL-15 measurement method are shown in Table 43.

[0290] Total IL- 15 (drug-bound and drug-unbound IL-15 / IL-15Ra) in serum was quantified using a ligand-binding assay. Both the drug-unbound free IL-15 / IL-15Ra and the drugbound IL-15 / IL-15Ra were first captured by a biotinylated anti-human IL-15 antibody that was non-competing with TEV-53408 on a Streptavidin-coated 96-well plate. Excess TEV-53408 was added to convert any free IL-15 / IL-15Ra to the bound form. Sulfo-tagged anti-idiotypic antibody against the TEV-53408 was then added to the washed plate for detection. By using the human IL- 15 specific capture / TEV-53408 specific detection antibody pair and a recombinant human IL- 15 / IL-15Ra calibrator, the assay measures total IL-15 / IL-15Ra receptor complex. Qualification results for the free IL- 15 measurement method are shown in Table 43.Table 43: Qualification of Free and Total IL-15 Measurement Assays%CV = percent coefficient of variation; LLOQ = lower limit of quantification; ULOQ = upper limit of quantificationExample 9: Clinical Pharmacology StudiesTable 44: List of Abbreviations for clinical pharmacology studies

[0291] TEV-53408 drug product was tested in a Phase 1 trial in healthy volunteers evaluating the safety, tolerability, pharmacokinetics, and immunogenicity of TEV-53408 following single dose and multiple doses. For doses of 50mg and above, 100 mg / mL TEV-53408 in 20 mM Histidine buffer, 250 mM Sucrose, 0.1 mM EDTA, 0.5 mg / mL Polysorbate 80 at pH 5.2) was used. For doses below 50mg, TEV-53408 was diluted to 20 mg / ml in the same buffer (20 mM Histidine buffer, 250 mM Sucrose, 0.1 mM EDTA, 0.5 mg / mL Polysorbate 80 at pH 5).

[0292] Participants were administered a single dose (SD) of 1, 5, 10, 50, or 150 mg; or a multiple dose (MD) (2.5 and 7.5 mg every 4 weeks [q4w]) subcutaneous (sc). The number of participants and the blood sampling schema (pharmacokinetic, pharmacodynamic, NK and ADA) for the study are summarized in Table 45 for both the SD and MD cohorts.Table 45: Single Dose and Multiple Dose Cohorts: Blood Samples CollectionADA=anti-drug antibody, IL-15=interleukin 15; MD=multiple dose; n=number; NK=natural killer; PD=pharmacodynamic; PK=pharmacokinetic; q4w=every 4 weeks; sc=subcutaneouslyaSD cohorts 3 to 5 were dosed in parallel.bMD cohorts received 3 total doses, single dose q4w.

[0293] A total of 79 participants were randomized to the 1, 5, 10, 50 and 150 mg SD cohorts, of which 60 participants received TEV-53408 (12 per dose cohort) and 19 participants received placebo. A total of 36 participants were randomized to the MD cohorts, of which 30 participants received TEV-53408 and 6 participants received placebo.TEV-53408 Serum Concentrations - Single Dose

[0294] Following SD administration of 1, 5, 10, 50, and 150 mg of TEV-53408, serum concentration of TEV-53408 increased with increasing dose. Peak concentrations were typically reached between 6 to 14 days. After reaching peak concentrations, TEV-53408 is eliminated with two identified slopes: a first elimination phase at concentrations above ~0.1 pg / mL, and a secondphase, with faster elimination at concentrations below ~0.1 pg / mL. In the 1 mg cohort, the pharmacokinetic profiles were predominately below ~0.1 pg / mL, thus elimination was governed mainly by the second elimination phase. In all other dose cohorts, especially the highest doses of 50 and 150 mg, the first half-life predominates the majority of the profile, and the portion of second elimination phase is negligible, thus first phase (above ~0.1 pg / mL) half-life is reported.

[0295] Mean concentration-time profiles from all SD cohorts until day 190 are shown in Figure 13 (linear scale) and until day 666 are shown in Figure 14 (semi-log scale). The TEV- 53408 pharmacokinetic parameters are summarized by cohort in Table 46.Table 46: Summary of TEV-53408 Serum Pharmacokinetic Parameters After Single Dose (Subcutaneous) by Cohort (Pharmacokinetic Analysis Set)a Arithmetic mean - unless otherwise stated.% AUC=percent of AUC extrapolated from time of the last measurable concentration to infinity; AUCo-t=area under the drug concentration-time curve from time 0 to the time of the last measurable drug concentration; AUCo-oo=area under the drug concentration-time curve from time 0 to infinity;CL / F=apparent clearance after sc dose; Cmax=maximum observed drug concentration; n=number; SD=single dose; t ' / 2=elimination half-life; tmax=time to maximum observed drug concentration; Vz / F=apparent volume of distribution after sc dose.TEV-53408 Serum Concentrations - Multiple Dose

[0296] In the MD1 cohort (2.5 mg) and MD2 cohort (7.5 mg), TEV-53408 was administered once q4w for a total of 3 administrations. TEV-53408 accumulation was evident in both cohorts on administrations of second and third doses (accumulation ratio [AR] = ~2) and steady state was not reached by the third administration on day 57 (Figures 15 and 16). All pharmacokinetic parameters for the MD cohorts are reported in Table 47. An increase in the elimination rate after the third administration was seen, as TEV-53408 concentrations declined to approximately 0.1 pg / mL. First elimination phase half-lives calculated for the third period up to day 246 are 23 and 31 days following the third administration for the 2.5 and 7.5 mg cohorts, respectively.Table 47: Summary of TEV-53408 Pharmacokinetic Parameters After Multiple Doses (Subcutaneous) by Cohort (Pharmacokinetic Analysis Set)aMean - unless otherwise stated.AUCtau=area under the concentration-time curve over 1 dosing interval; AUCo- / =area under the concentration-time curve from time 0 to infinity; Cmax=maximum observed drug concentration; Ctrough=trough concentration; MD=multiple dose; n=number; N / A=not applicable; t / 2=elimination half-life; tmax=time to maximum observed drug concentration.Dose Proportionality

[0297] The single-dose proportionality of TEV-53408 was evaluated over the dose range of 1 through 150 mg and for the range of 5 through 150 mg using a power model based on maximum observed drug concentration (Cmax), area under the concentration time curve from time 0 to the time of the last measurable concentration (AUCO-t), and area under the concentration-time curve from time 0 to infinity of the last measurable concentration (AUCO-co). For the dose range of 1 to 150 mg, increases in exposure were greater than dose proportional. For the dose range of 5 to 150 mg, increase in Cmax was dose proportional; however, increases in AUCO-t and AUC0-co were greater than dose proportional.Free Serum IL- 15 - Single Dose

[0298] Target engagement of TEV-53408 was evaluated using free serum IL- 15 levels (not drug bound). The assay measures free IL-15 / IL-15 receptor alpha chain (IL-15Ra) complex. Baseline IL- 15 concentrations for all healthy volunteer SD cohorts were around 5 ng / L. A dosedependent effect of TEV-53408 was observed on free serum IL-15 (Figure 17 and Figure 18). In all SD cohorts, a rapid decline in free IL-15 levels occurred in the first day following administration of TEV-53408. In all cohorts except 1 mg, IL-15 was neutralized to below the limit of quantification (BLQ), with a lower limit of quantification (LLOQ) of 2.4 pg / mL. The duration period of BLQ values increased with increasing dose. Following the initial decline, IL-15 concentrations rose until they reached a peak approximately 3 to 4 times above baseline levels before returning to baseline (or close to baseline). Time to peak and time back to baseline also increased with increasing dose. Peak concentrations of IL- 15, however, were slightly lower with higher doses compared to lower doses (Figure 17 and Figure 18). Note: Free IL-15 is IL-15 / IL- 15Ra complex.Free Serum IL- 15 - Multiple Dose

[0299] The mean free IL- 15 concentration versus time by treatment is presented in Figures 19A-19C. For both healthy volunteer MD dose levels, free IL-15 concentrations declined after administration of TEV-53408; however, the extent and duration of decline varied between the doses, indicating a dose dependent effect.Total Serum IL- 15 - Single Dose

[0300] The mean total IL- 15 concentration versus time by treatment is presented in Figure20. In all SD cohorts, an initial rise in total (free and drug-bound) serum IL- 15 was observed. Total IL- 15 increased in a dose-dependent manner in all SD cohorts, in terms of extent and duration.Total Serum IL- 15 - Multiple Dose

[0301] Total IL-15 increased with each administration in both cohorts. Mean concentrations in the 2.5 mg cohort returned to baseline by day 204 while concentrations in the 7.5 mg cohort, although declining, remained above baseline (Figures 21A-21B).TEV-53408 causes a decline in NK cell numbers

[0302] IL- 15 has been shown to stimulate the proliferation, activation, and survival of natural killer (NK) cells, intraepithelial lymphocytes (IEL) and other lymphoid cells and is postulated to play a role in mediating immune diseases. A previous study showed that anti -IL- 15 antibody AMG 714 did not cause a reduction of NK cells in human clinical trials (Lebrec H et al. J. Immunol. 2013; 191(11):5551-8). Lebrec concluded that IL- 15 was not essential for NK cell survival in humans due to cytokine redundancy. Multiple cytokines can stimulate NK cell maturation, proliferation and activated effector function:. Including type I IFN, IL-12, IL-18 and IL-15 (Walzer, T et al., Natural-killer cells and dendritic cells: ‘Tunion fait la force.” Blood 106, 2252-2258 (2005).

[0303] Surprisingly, in all SD cohorts except 1 mg, a decline in NK cell counts occurred in all participants receiving TEV-53408 within the first 2 weeks after drug administration. This decline recovered with time in all dose levels. Subjects at the highest doses had NK cell suppression for over 6 months. (Figure 22). This decline was confirmed in the relevant geographic and temporal population by measuring the NK cell population for all 115 participants at day 1, and following placebo subjects for at least 6 months.Table 48: NK cells as % of circulating PBMC for untreated participants.

[0304] The extent of NK cell count decline wa's dose-dependent, with counts declining to a lesser extent in the 5 mg cohort than in 10, 50 and 150 cohorts. With doses of 10 mg and higher, NK cell counts declined by approximately 70% from baseline, despite the difference in doses, indicating the effect plateauing. The duration of decline increased with increasing doses. During the time of decline, mean NK cell count remained above 70 cells / pL and above the lower limit of normal (LLN; 24 cells / pL or 1.4% of total blood lymphocytes). A limit of 1.4% was based on the lowest validated sensitivity of commercial pathology lab tests. Expert opinion considers 1% as more relevant number based on observations of those with serious NK defects, where risks are mitigated when NK cells are at least 1% of PBMC and are functional, (Orange J Allergy Clin Immunol. 2013 Sep;132(3):515-525; Orange JS.J Clin Immunol. 2020 Jan;40(l): 13-23). Subsequent studies have validated the commercial assay to 1% of lymphocytes, and this will be the LLN for future studies.

[0305] It is surprising to see increased dose affected the period of NK cell count suppression more than the degree of suppression, with a plateau or “floor effect” limiting the maximal suppression.

[0306] Five of the 24 (21%) participants in the 50 and 150 mg cohorts had temporary excursions below the value of the central laboratory’s LLN of 24 cells / pL, however, no adverse events occurred in these participants. All but one of the events had declined below the LLN in the 50 and 150 mg cohorts were isolated, with values ranging between 15 to 22 cells / pL (Figure 23).

[0307] A participant in the 150 mg cohort had several readings below the LLN: 3 consecutive readings over approximately 4 weeks (19 cells / pL on day 29, 12 cells / pL on day 43, 14 cells / pL on day 57) and 1 isolated reading on day 190 (17 cells / pL). On day 638 (last measurement available for this participant), NK cell counts were 212 cells / pL. No treatment emergent adverse events were reported for this participant.

[0308] A dose-dependent effect on mean NK cell counts was also observed in the MD cohorts (Figures 24A-24B). In the 7.5 mg cohort, a decline in NK cell counts was observed after first administration, similar in magnitude to that of the 5 mg SD cohort (Figures 24A-24B). The second and third administrations of TEV-53408 did not cause further decline, and cell countremained stable up to day 246, with mean NK cell count above 90 cells / pL. No declines below LLN were observed in MD cohorts.

[0309] This result was further surprising in view of studies in monkeys and mice, where TEV-53408 lowered NK cells below detectable levels.After treatment, circulating NK cells retain cytotoxic function.

[0310] To determine the normal characteristics of the population, day 1 (predose) samples were examined, and the blood from placebo controls were obtained over a 6 month period. Samples were analyzed for NK cell numbers, Mean percentage of CD56dim / bright(Table 49), CD 107a expression and perforin both before and after stimulation with phorbol myristate acetate (PMA) / ionomycin to induce degranulation.Table 49: Mean percentage of CD56dim / brightin 81 untreated subjects.Note: sd = standard deviation

[0311] 25 placebo controls were also followed for NK cell function for a 6 month period to determine natural variation. % NK cells and CD56dim / brightvaried considerably. Figure 27 shows the % of CD56 bright cells over 204 days in placebo controls, showing the extent of natural variation in CD56-positive cell numbers.

[0312] The range of CD 107a expression and perforin content by CD56dim NK cells after PMA / I stimulation, between cryopreserved PBMCs from healthy donors, is wide at day 1 (Table 50) and also showed intra-individual variability over time (not shown).Table 50: Range of NK cell activation markers in 81 untreated subjects.

[0313] In TEV-53408-treated groups samples from participants in single dose (SD) cohorts (1, 5, and 10 mg) at 5 different timepoints from baseline up to day 134 and at 8 different timepoints from baseline to day 204 in the multiple dose (MD) cohorts (2.5 and 7.5 mg) were further assessed for NK cell function. The distribution of maturity (CD56brightand %CD56dimat all timepoints incirculating NK cells) ranged from 83-94%, throughout the trial and at nadir levels. This range overlapped with the 88-97% range (min-max) in placebo group, and was strongly affected by outliers.

[0314] Figures 28A-28F show the distribution of CD107a expression before and after in- vitro stimulation by phorbol 12-myristate 13-acetate (PMA) / ionomycin, for NK cells from individuals treated with a single TEV-53408 dose of Img (Figure 28A), 5 mg (Figure 28B), or lOmg (Figure 28C), with multiple TEV-53408 doses of 2.5mg (Figure 28D), 7.5mg (Figure 28E), or placebo (Figure 28F). The degranulation ability (shown by CD107a expression) of NK cells was not affected by different doses of TEV-53408, multiple doses of TEV-53408, or over time. All NK cell samples at all doses preserved the ability to release perforin after stimulation (data not shown).

[0315] The data shows that circulating NK cells in TEV-53408-treated individuals retain cytotoxic functions. This is consistent with the lack of serious adverse effects, including no increase in viral infection or malignancy (below).Immunogenicity

[0316] Blood samples were collected for the analysis of anti-TEV-53408 antibodies predose on day 1 (of each dose) and day 15 for all cohorts. ADA samples for each of the SD and MD cohorts were collected per the study protocol (see Table 45 for details). A participant was classified as having a treatment-emergent ADA response if either: i) a participant had an ADApositive sample at any of the post-dose time points, but not at the pre-dose time points or ii) a participant had ADA-positive samples at pre-dose time points and 1 or more at post-dose time points with at least a 4-fold increase in post-dose titers relative to the pre-dose titers.

[0317] Treatment-emergent ADA responses were observed in 3 of the 60 participants who received TEV-53408 in the SD cohorts. One participant had 3 sporadic ADA-positive samples throughout the study, and the other 2 participants had ADA-positive samples from day 85 until day 470. There were no events of hypersensitivity, or any other adverse events temporally associated with the ADA-positive participants. TEV-53408 concentrations as well as the effect on free IL-15 levels and NK cell counts of the ADA-positive participants were similar to those of the ADA negative participants.

[0318] In samples through day 134 of the MD treatment period four of the 30 participants had ADA positive samples. Three of the 4 participants had ADA-positive samples at baseline (prior to dosing of TEV-53408) and post-dose time points but did not have a 4-fold increase in their post-dose titers, and therefore, did not meet the criteria for a treatment-emergent ADA response. Another participant had only one ADA titer (on day 85); the transient ADA response in this participant was not associated with any safety signal. There were no events of hypersensitivity, or any other adverse events temporally associated with the ADA-positive participants. TEV-53408 concentrations as well as the effect on free IL- 15 levels and NK cell counts of the ADA-positive participants were similar to those of the ADA-negative participants.Lack of Adverse Effects

[0319] The functional consequences of NK cell deficiency are unknown. Traditionally, inborn errors of immunity (lEIs) involving NK cell deficiency have been felt to be associated with various infections. However, these lEIs are typically associated with deficits in other aspects of immunity, making it difficult to ascertain the specific role of NK cells. Much of the published data available regarding Individuals with NK cell IEI usually present with severe, recurrent, or atypical herpes viral (cytomegalovirus [CMV], herpes simplex virus [HSV], varicella-zoster virus [VZV], and Epstein-Barr virus [EBV]) or human papillomavirus (HPV) infections that are not lifethreatening. In regard to malignancy, some with the GATA2 deficiency (a more general lymphoid cell deficiency syndrome) appear to have an increased risk of HPV-associated cancers, yet the experiential and published data on the association of other forms of NK cell lEIs and malignancy is scant. Vely et al. described a cohort of patients with severe combined immunodeficiency who underwent hematopoietic stem cell transplant and had long term, selective deficiencies in NK cells and other innate lymphoid subsets (Vely et al, Nat. Immunol. 2016, PMID 27618553). These patients were not found to have an increased rate of infection. While Vely’ s data suggests that NK cells and other ILCs can be redundant in normal protective immunity, it is premature to extrapolate to a decline in NK cells caused by blocking IL-15.

[0320] Accordingly, participants were closely monitored for viral infection or cancer. All participants were monitored for adverse events, vital signs, ECGm serum chemistry, hematology, coagulation and urinalysis.

[0321] In the SD cohort, TEV-53408 was well tolerated after SD administration in healthy participants in the dose range of 1 to 150 mg. 17 / 60 (28%) participants from the TEV-53408 group and 6 / 19 (32%) participants from the placebo group reported at least 1 adverse event. All adverse events were considered not related to IMP. One participant from the 50 mg cohort reported a spontaneous abortion that was considered unrelated to TEV-53408 treatment. One participant exhibited elevated ALT following TEV-53408 dosage, but also concomitantly with exposure tochemicals and alcohol consumption such that the cause could not be determined. One participant from the 150 mg cohort reported a mild herpes zoster infection 13 months after dosage. This was considered not related to treatment. No systemic severe hypersensitivity reactions (including anaphylaxis) or malignancies were reported in the SD cohorts during the study.

[0322] In the MD cohort, TEV-53408 was well tolerated after MD administration in healthy participants in the dose range of 2.5 mg and 7.5 mg q4w. There were no deaths in the study. 1 / 30 (3%) participant from the TEV-53408 group and 1 / 6 (17%) participant from the placebo group reported at least 1 adverse event. Both the events were considered as not treatment-related.

[0323] The study took place during the COVID outbreak, in Florida, yet there were no increases in COVID infection or parasitic infections from the TEV-53408 group.

[0324] Extended follow up to the multiple dose cohort demonstrated a return of NK cell numbers to baseline, and a lack of serious adverse events or adverse reactions linked to the drug. Plots of IL-15, antibody and NK cell levels are provided in Figures 19C, 21B and 24B. A summary of pK parameters from single dose and multiple trials is provided in Table 51.Table 51: Summary of PK parameters following SD or MD administration of TEV-53408

[0325] aArithmetic mean (standard deviation);bAUCo- / for SD, AUC for MD;cMedian(range);dn=l 1;en=10;fParameters are reported after the last dose;gn=13;hn=12. AUCo-®, area under the serum concentration-time curve extrapolated to infinity; AUC , area under the concentration-time curve over 3rd dosing interval; Cmax, maximum observed serum drugconcentration; MD, multiple dose; q4w, every 4 weeks; SD, single dose; tmax, time to maximum serum concentration; ti / 2, half-life. No serious adverse effects were observed.

[0326] The extended MD study and further modeling confirm previous models.Identification of treatment dose

[0327] Three pharmacometric models were developed based on the data in healthy volunteers: A population pharmacokinetic model, a pharmacokinetic pharmacodynamic model to describe the relationship between exposure and effect on free serum IL-15, and a pharmacokinetic- pharmacodynamic model to describe the relationship between exposure and effect on NK cell counts.

[0328] Pharmacokinetics of TEV-53408 in healthy volunteers was described by a two- compartment population pharmacokinetic model, with combined linear and saturable elimination, first-order absorption, and weight as a covariate on clearance and volumes of distribution. The model was used in simulation mode to anticipate TEV-53408 exposures with alternative doses and dosing regimens and to simulate exposure (e.g. 10 q4w, 50 qw4, 50 ql2w). A model of exposureon administration of 150 mg ql2w on 2 administrations is shown in Figure 25.

[0329] The relationship between TEV-53408 exposure and the effect on free IL-15 in healthy volunteers was described by a joint pharmacokinetic-pharmacodynamic model. The model included standard target-binding dynamics and additional feedback mechanism to capture the apparent rebound of free IL-15 observed in clinical data. The model simulations were run to anticipate scenarios with repeated administration of the same dose levels already tested in FIH trial.

[0330] The relationship between TEV-53408 exposure and the effect on NK cell count in healthy volunteers was described by a joint pharmacokinetic-pharmacodynamic model. Indirect response model was used as a basis of the model structure. The model was used in simulation mode to predict the magnitude and duration of reduction in NK cell count with alternative doses and dosing scenarios for subsequent studies (Figure 26).Dose Rationale for Celiac Disease Treatment

[0331] Animal models, as summarized in Example 20, support a range of doses. No serious adverse health effects were observed in mice dosed up to 2.5mg / kg of body weight or in monkeys up to 400mg / kg. Effectiveness in a murine model of celiac disease was observed at a dose of Img / kg, and in monkeys at lOmg / kg. These correspond to human doses of 70mg and 700 mg.

[0332] The clinical trial data showed that TEV-53408 well tolerated over a large range of doses without any serious reactions or adverse events. The observed drop in serum NK numbers led to a revised guidance to determine the dose for treatment of celiac disease. The following quantitative guiding principles were taken into consideration.• A dose sufficient for suppression of circulating free IL- 15 during dosing interval.• That the PK parameters not exceed highest exposures obtained following 150 mg single dose (Cmax,SS <= Cmax of 150 mg SD and AUCSS <= AUCinf of 150 mg SD).• Ensuring NK cells did not fall below lower limit of normal.

[0333] Based on the clinical trial, dosing regimens selected for the treatment of celiac disease, including non-responsive celiac disease, were between 10 mg-50mg antibody every 4 weeks, and between 50-150mg antibody every 12 weeks.

[0334] More specifically10 mg TEV-53408 every 4 weeks (10 q4w) ;20-30 mg TEV-53408 every 4 weeks (20-30 q4w);50 mg TEV-53408 every 4 (50 q4w); and150 mg TEV-53408 every 12 weeks (ql2w).

[0335] The TEV-53408 formulation will be 100 mg / ml (for 50 mg and 150 mg doses) or 20mg / ml (for lower doses) in 20 mM Histidine buffer, 250 mM Sucrose, 0.1 mM EDTA, 0.5 mg / mL Polysorbate 80 at pH 5.

[0336] Ten milligram q4w is considered a minimal anticipated therapeutic dose based on the noticeable effect on NK cells (refer to observed and simulated data). Both 50 mg q4w and 150 mg ql2w are anticipated to be efficacious based on sustained reduction of IL- 15 during the dosing interval. 150 mg ql2w is expected to result in a similar Caverage as 50 mg q4w, but permit exploration of a correlation of Cmax and / or Ctrough to efficacy / safetyTEV-53408 10 mg and 50 q4w are anticipated to result in separated exposures (5 fold difference) to adequately inform the E / R model (for efficacy and safety). The guiding principle for dose selection, is the anticipated different effect on circulating IL- 15 reduction during the dosing interval (from none to complete reduction). The proposed dosing regimen is expected to induce a sustained decrease on free serum IL- 15 during treatment.

[0337] Following review of the entirety of the data available to date from Trial there are no clinical or laboratory data (including the observed reduction in NK cells) that raise clinical concern for the dosing regimens. No safety concerns or adverse events related to NK cells declinewere identified at doses of 1 to 150 mg following single dosing and at 2.5 and 7.5 q4w for a total of 3 administrations in healthy participants.

[0338] It is noted that correlation of these measures to effect in the gut is unknown at in humans. Preclinical data herein demonstrate that doses reducing serum IL-15 and NK cells are able to reduce Tim activity at the site of disease and show a treatment effect. The effect of TEV- 53408 at different exposures will model identification of optimal dose / frequency.Treatment of Vitiligo with TEV-53408

[0339] Vitiligo is an autoimmune disease that causes depigmentation of the skin. There is no cure for vitiligo, and approximately 40% of patients relapse within one year of stopping treatment. Vitiligo is believed to be mediated by IL-15 dependent skin-resident CD8+ T-cells that target melanocytes, analogous to the role of gut-resident CD8+ T-cells responding to gluten in celiac disease.

[0340] TEV-53408 was shown to be effective in a mouse model of vitiligo disease, suppressing melanocyte-specific CD8 T cells (see below). Dosing above Img / kg was found to be effective in this model. Accordingly, treatment in humans could be between about 10 mg to about 50mg antibody every 4 weeks, and / or between about 50 mg to about 150mg antibody every 12 weeks.

[0341] More specifically about 10 mg TEV-53408 every 4 weeks (10 q4w) ; about 20mg TEV-53408 every 4 weeks (20 q4w); about 30 mg TEV-53408 every 4 weeks (30 q4w); about 50 mg TEV-53408 every 4 weeks (50 q4w); about 50 mg TEV-53408 every 12 weeks (50 ql2w); and about 150 mg TEV-53408 every 12 weeks (ql2w).

[0342] A dosing regimen of TEV-53408 150 mg ql2w for a total of 2 administrations was selected to explore safety and effect of TEV-53408 in vitiligo patients. This dosing regimen is supported by the available data shown above, with 2 years of follow up for the 50 mg and 150 mg cohorts. A dosing regimen of 150 mg ql2w is considered to have a favorable administration frequency and alleviate patients' burden.

[0343] With the proposed dosing regimen, TEV-53408 is expected to induce a sustained decrease on free serum IL- 15 during the treatment period (24 weeks). Suppression of IL- 15 isanticipated to suppress activity of Trms within the skin and allow sufficient time for proliferation, migration, and differentiation of new melanocytes within lesions.

[0344] An increase in free serum IL- 15 levels occurring after initial reduction was observed following single administrations of TEV-53408 at 1 to 150 mg. In the MD cohorts, the first administration elicited a reduction in free IL- 15 for a mean period of 7 days. Following the consecutive second and third doses, a lower extent of reduction in free IL- 15 was observed..

[0345] As described above, a population pharmacokinetic model was used to simulate the anticipated exposure to TEV-53408 on administration of 150 mg ql2w (Figure 25). Based on the simulations and the mean half-life of TEV-53408 150 mg following SD administration (38 days), accumulation of 31% in Cmax and 32% in AUCtau after the second dose is expected (Figure 25 andTable 52)Table 52: Anticipated Exposure (Cmax and AUC) of 150 mg q!2w for 2 DosesAUC=area under the curve; AUCo- / =area under the drug concentration-time curve from time 0 to infinity; Cmax=maximurn observed drug concentration; q!2w=every 12 weeks

[0346] In general, a dosing regimen of 150 mg TEV-53408 ql2w is modeled to provide optimal clinical effectiveness based on the length of IL-15 neutralization observed in serum, NK cell suppression, as well as an expected benign safety profile observed in human clinical trials. Moreover, both the route and frequency of administration should support high patient compliance in this patient population.Example 10: Human bone marrow on a chip model of effect of TEV-53408 on NK cells

[0347] As an adjunct to the clinical data derived from dosing healthy human subjects with TEV-53408, we sought to model the effect of TEV-53408 on bone marrow using an “organ on a chip” model.

[0348] Humanized and immunocompetent organ-on-a-chip model combining a neuroblastoma cell line with primary human NK-cells isolated from PBMCs was described byMarzagalli M, et al (A multi-organ-on-chip to recapitulate the infiltration and the cytotoxic activity of circulating NK cells in 3D matrix-based tumor model. Front Bioeng Biotechnol. 2022 Jul 25;10).

[0349] The HUMIMIC Chip2™ (TISSUSE GMBH), contains a zirconium oxide ceramic scaffold with a hydroxyapatite coating to mimic the architecture of the bone marrow, and is supplied by microfluidic circuits to exchange medium. The HUMIMIC Chip2™ was seeded with 300,000 human mesenchymal stromal cells (MSC) serving as supporting cells and allowed to grow for 9 days to enable distribution over the scaffold surface, production of extracellular matrix and induction of an osteogenic phenotype. 40,000 presorted CD34+ hematopoietic stem and progenitor cells (HSPC) from three different donors were seeded onto the stromal cell covered scaffold in separate chips, in duplicates and transferred into the microfluidic system on the day the assay was initiated by adding a hematopoietic cell growth medium containing TPO, FLT-3L and SCF for HSC maintenance and IL-7 and IL-15 for NK cell differentiation. Initial cell pools of CD34+ cells on day 0 after thawing contained considerable amounts Stage 1 NK-cells (5.92%, 14.9% and 29.3%) and low amounts of Stage 2A and Stage 2B NK-cell progenitors.

[0350] Over time HSPCs and differentiated cells were transported by the microfluidic circulation into the medium compartment where they accumulated. The number of circulating cells sedimenting in the medium compartment increased over culture time and consistently reached close to 100% confluency. Those cells were sampled weekly from the system and used for counting and flow cytometry analysis on day 7, 14, 21, 28 and 35 of the assays. Culture conditions allowed cultivation over 7 weeks. Cells were analysed with a flow cytometry panel containing the early progenitor marker CD34, the lymphoid progenitor marker CD 10, the Interleukin-7 receptor- a (CD 127), the Killer cell lectin-like receptor subfamily B, member 1 (CD 161), the prototypic NK cells marker (CD56), type III Fey receptor (CD16), Interleukin-2 receptor subunit beta (CD122), the myeloid marker CD33 and the leukocyte common antigen CD45.

[0351] Expression patterns of NK-lineage cells in the primary bone marrow-derived mononuclear cells matched for the most part with expression patterns described in the literature.

[0352] Harvested cell counts from the medium compartment increased weekly and all three tested donors had a similar cell output. Major populations in the analyzed cells throughout the experiment were non-lymphoid HSPCs, myeloid cells, and a smaller fractions of NK progenitor cells. On average, the cultures contained around 0.91% NK-cells and 3.23% dendritic cells, from which the majority were inflammatory dendritic cells. The most prominent cell populations were CD14+ cells labeled as monocytes and CD66b+ granulocytes.

[0353] Differentiated NK cells could also be activated. Circulating cell pools were stimulated by Phorbol-12-myristat-13-acetate (PMA) and ionomycin to induce lysosomal- associated membrane protein- 1 (CD 107a) expression on the cell surface, which is a marker for NK cell activation and degranulation. Upregulation of CD 107a cell surface expression upon PMA / ionomycin stimulation was detected on all populations with increasing expression strength from Stage 2B to Stage 5.IL-15 removal depresses NK-cell development

[0354] IL- 15 -I- mice lack NK cells and exhibit reduced NKT cells (Carson et al 1997. J. Clin. Invest. 99:937-943; Kennedy MK et al. J Exp Med. 2000 Mar 6; 191 (5):771 -80) and other studies have shown the role of IL-15 in NK cell development in mice. To understand the role of IL- 15 in the HUMIMIC system, IL- 15 was withdrawn from the chips after day 28. The counts of stage 1 to 3 were not affected by removal of IL- 15 from the culture medium. A decrease in the NK cell counts of Stages 4 and 5 NK was detected both in circulation as well as in the ceramic scaffold compared to circuits with continuous IL- 15 supplementation (Figure 29A). This is consistent with the reported expression patterns of CD 122 (Abel et al (2018). Front. Immunol 0: 1869; Wang X and Zhao X-Y (2021). Front. Immunol. 12:610789. doi: 10.3389 / fimmu.2021.610789).

[0355] As shown in Figure 29B, sampled cell counts of the lymphoid / NK-cell lineage from the circulation over a timeframe of 49 days varied, with Stage 4 NK cells increasing through day 49. Figure 29C shows exemplary flow cytometry plots of CD56 and CD 16 expression in CD161+ cells in the circulation and in the ceramic scaffold, used for gating of Stage 3, Stage 4 and Stage 5 NK-cell populations on day 49 of the assay, with an enrichment of cells visible in Stage 4 in circulation, but transition to stage 5 observed in the ceramic scaffold.

[0356] Figure 29D shows the fraction of CD16+ cells (Stage 5 NK-cells) of all CD161+ CD56+ NK-cells at day 35 and day 49 sampled from the circulation or harvested from the ceramic scaffold. At both day 35 and day 49, the proportion of CD161+ CD56+ NK-cells was increased.

[0357] Figure 29E shows an exemplary flow cytometry plot of CD 107a expression on Stage 4 NK-cells with and without PMA / ionomycin stimulation. An increase in CD107a-BV421 staining was observed following stimulation.

[0358] Figure 29F shows the mean fluorescence intensity of CD107a-BV421 on Stage 2B, Stage 4 and Stage 5 NK-cells at day 35 with and without PMA / ionomycin stimulation. IncreasedMFI of CD107a-BV421 was observed across all three stages following PMA / ionomycin stimulation.

[0359] Surprisingly, IL-15 removal only abrogated Stage 4 and 5 by less than one log. and there was marked donor-donor variation. Donor 2 only showed minor changes upon IL- 15 removal while cells derived from donor 1 were strongly decreased in circuits without constant IL- 15 supplementation. This suggests that IL-15 is being expressed by cells in the chip and can be providing IL- 15 in soluble form or through cell-cell signaling, so called “trans presentation”. Dendritic cells have been shown to express IL-15 (Dubois et al Immunity, 17(5), 2002, pp 537- 547).Anti IL-15 antibody induces a response similar to IL-15 removal

[0360] The HUMIMIC Chip2™ system was treated for one week or two weeks with 250 pg / mL of monoclonal antibody against IL- 15, starting either at day 21 or day 28 of the assay. IgG matching isotype control at the same concentration of 250 pg / mL was used as negative control.

[0361] At day 35, a significant reduction in Stage 4 and Stage 5 cell counts in the ceramic scaffold (Figure 30B) and the circulation (Figure 30A) in comparison to the IgG control was measured for all donors. Treatment effects were visible for chips treated for either one (day 28 - day 35) or two weeks (day 21 - day 35). All other NK-cell progenitor populations were not affected by the application of TEV-53408.

[0362] CD34+ cell donor 2 and an additional donor 4 were used in two additional experiments to analyze dose-dependency of the antibody effect in a range of 0.25 pg / mL, 2.5 pg / mL, 25 pg / mL up to 250 pg / mL. Treatment with TEV-53408 was initiated at day 28, since the effect was similar between the two tested timepoints, and higher counts of stage 4 NK cells were detected on day 28. Cells were sampled from the medium compartment on day 28, day 31 and day 35 of the assay in the experiment with donor 2 and at day 28 and day 35 in the experiment with donor 4.

[0363] Concentration dependent treatment effects were visible on stage 4 NK-cells at antibody concentrations of 2.5 pg / mL and higher after seven days of treatment. To account for underlying variability of NK-cell counts at day 28, Stage 4 NK-cells counts at day 31 and day 35 were normalized to the baseline counts of the same circuit at day 28 for calculation of the IC50. Dose-dependent reduction of Stage 4 NK-cell numbers in circulation were detected after seven days of treatment with TEV-53408 for both tested donors. The difference between calculated IC50 values of both donors was not significant (Figure 30B). The dose-dependent treatment effect ofthe antibody increased over time as indicated by a lower IC50 value on day 35 (1.91 pg / ml) compared to day 31 (21.49 pg / ml) for donor 2 (Figure 30B).

[0364] Frequency of EdU+ proliferating cells in the Stage 4 NK-cell population was measured in circulation samples on day 28, day 31 and day 35 of the assay. The baseline frequency of proliferating Stage 4 cells on day 28 was around 15%. The proliferation rate of Stage 4 NK cells was significantly reduced on day 31 upon treatment with TEV-53408 concentrations of 2.5 pg / mL or higher to below 5% proliferation rate at the highest concentration. On day 35 higher proliferation frequencies were measured compared to day 31 but still significantly lower than in the control condition (Figure 30C). Total numbers of Stage 4 NK cells were reduced, consistent with the numbers of proliferating Stage 4 NK cells on day 35. Treatment-induced reduction of the proliferation rate was not observed in the Stage 2B progenitor population. This result is consistent with the role of IL- 15 in NK cell maturation and proliferation.

[0365] To show drug reversibility, several chip circuits of HSC donor 4 were supplemented with fresh differentiation medium after the treatment with TEV-53408 was stopped at day 35, and the differentiation of NK cell stages was traced for two additional weeks (Figure 30D). Reversibility was demonstrated in a dose-dependent manner, with circuits treated with lower concentrations of antibody recovering faster after removal than those circuits treated with higher TEV-53408 dose, in both circulating (Figure 30E) and in the ceramic scaffold (Figure 30F). It is unclear if slower recovery is due to residual TEV-53408 remaining in the circuits.

[0366] Mean fluorescence intensity (MFI) of CD 107a Stage 4 NK cells was measured with and without PMA / ionomycin stimulation at day 35 of the assay after 7 days of treatment with TEV- 53408 and during the recovery period. Strong activation with almost lOOx increase of the MFI was observed on all timepoints from day 35 to day 49. No differences were detected between cells that were treated with the TEV-53408 and untreated cells indicating that the treatment did not affect the ability of the remaining Stage 4 NK cells to be activated (Figure 30G).

[0367] The effect of removing IL-15 on NK cells in the HUMIMIC chip model is similar to TEV-53408. TEV-53408 also appears to produce effects on NK cells that are similar in both the HUMIMIC model and clinical data. There is a reduction of but not complete abrogation in mature circulating NK cells: a less than 1 log decline on average, even with increasing dose. There is also a recovery in NK cell production after removal of the TEV-53408, although the time to recovery increases with increasing TEV-53408 dose. In both systems, remaining circulating NK cells retain the ability to be activated.

[0368] Compared to removing and then restoring IL- 15 in the HUMIMIC system, the delay in NK cell recovery after removal of TEV-53408 in the HUMIMIC system was slower and dose dependent. This is most easily explained as due to persisting TEV-53408. It is also possible, however, that the TEV-53408 affects cells in ways that are more than simply explained by loss of IL-15. In the clinical trial, NK cell levels returned to baseline levels in a dose dependent manner and long after IL- 15 levels returned to normal and increased above baseline. This is surprising..

[0369] TEV-53408 reduces NK cells only partially, and reversibly, and does not abrogate mature NK cell function, and participants in the clinical trial showed no increase in viral infection or drug linked serious side effects in the participants. This appears distinct from

[0370] the extreme deficiencies in NK cells that have been associated with increased risk of specific viral infections and cancers, such data is from individuals with NK cells well below those observed after treatment with TEV-53408, and typically in people with more comprehensive immune defects. .Example 11: Increase in IL-15 levels with declining antibodies is not associated with adverse events.

[0371] While elevated levels of IL- 15 are typically linked to certain inflammatory disease, there does not appear to be a universally accepted definition of “normal” or what is “elevated”. According to in vitro studies, more than 10 nM IL- 15 was needed to induce meaningful proliferative responses in NK cells. (Kobayashi Blood (2005) 105 (2): 721-727). We have observed a 15pM IC50 for the cell line NK-92 (data not shown). Others have reported that NK- 92 proliferates in response to -0.25 ng / mL of exogenous IL-15 (free) (19.3 pM) (Tomroos, et al. ANTICANCER RESEARCH 39: 107-112 (2019)).

[0372] At the other extreme, administration of 3pg / kg / day of IL-15 to cancer patients caused serum levels to peak at 1,260 pg / ml and was followed by upregulation of inflammatory cytokines, increased NK cells >10x, CD8 T-cells 8x, and CD4 T-cells 3x. (Conlon et al, J. Clin Oncol 2015 Jan 1 ;33 (1): 74-82). Some patients reported symptoms consistent with acute cytokine toxicity id.).

[0373] In vivo, the relationship between IL-15 levels and disease is complex. Because IL- 15 signalling can occur between cells (trans), serum or tissue levels do not fully reflect changed IL- 15 levels. Different measuring methodologies also add variation, such that a relative measure can be more informative within a specific situation.

[0374] Heydari et al (J. Gastrointestin Liver Dis, September 2018 Vol. 27 No 3: 241-247) found higher mean serum levels of IL-15 in patients with celiac disease (69.4 ± 137.9) as compared with patients with non-celiac gluten sensitivity (27.9 ± 61.1) and control (17.0 ± 43.9) group, but these differences were not significantly different between the studied populations (p= 0.869).

[0375] Atwa et al (J Cosmet Dermatol 2021 Aug;20(8):2640-2644), reported that controls had 26.4 pg / ml in serum while those with Vitiligo were 45.67 pg / ml in serum, but that the elevated level correlates with disease severity, but not activity. Chen et al. (Free Radio Biol Med 2019 Aug 1 : 139:80-91) reported lOpg / ml in controls and 15 pg / ml in patients, with correlation to VASI score. Kassab A, et al (J Clin Med. 2023 Sep 9; 12(18):5861. showing 80-100% increase in IL-15 levels in vitiligo patients vs. controls (depending if they are active or stable):

[0376] Significant differences between controls (60 pg / ml) and alopecia areata patients (90 pg / ml) (Tabara K et al. Advances in Dermatology and Allergology 2019;36(l):63-69). Ragab et al reported serum level of IL-15 was 13.20-41 pg / ml (mean ± SD: 22.80 ±5.99 pg / ml) in patients, while in the controls was 0-5.9 pg / ml (mean ± SD: 0.3 ±1.32 pg / ml).

[0377] We measured approximately 5 ng / L of free IL- 15 as the baseline level in healthy human subjects and this level was maintained in the placebo group. Administration of TEV-53408 decreased serum free IL- 15 to below the level of detection for a period dependent on the antibody dose. Subsequent doses rose to as much as 4 times that prior to TEV-53408 administration, including over 20 ng / L. Peak levels of free IL- 15 in those administered a single dose of 50 or 150 mg were approximately 15ng / L, while those administered a single dose of 1, 5, or 10 mg peaked at an average of 22 ng / L, and those administered multiple doses of 2.5 or 7.5 mg peaked at an average of 17 ng / L. These levels of free IL- 15 are below the amounts believed to be sufficient to trigger IL-15 mediated responses.

[0378] It is acknowledged that the levels in blood may not fully reflect levels in tissue, or the amount of trans signalling available. Nevertheless, in contrast to publications reporting a link to disease when levels were elevated by less than 2x, elevated IL-15 levels following TEV-53408 administration were not associated with adverse events or measurable increases in serum markers for inflammation or liver toxicity.

[0379] Should elevated IL- 15 be of concern, the pK data provide a strategy to mitigate the rise in serum IL-15 following TEV-53408 administration. Figure 19C shows a direct comparison of MD 7.5mg and SD 5 and lOmg. To the extent that thecontinued presence of TEV-53408 can produce a lower peak IL- 15 level, a lower peak IL- 15 level can be obtained by tapering off theadministered dose. For example, following multiple doses of 50 or 150 mg / ql2w, a patient can be administered one or more doses of 7.5 mg.Example 12: TEV-53408 in knock-in mice reflects that in other modelsConstruction of mouse

[0380] A knock-in mouse was generated on a C57Black / 6 background, deleting the murine IL15 gene and adding a cassette encoding codon optimized sequences for the complete human IL- 15 and human ZL-15Ra, with an IRES sequence there between. These mice are abbreviated hlL- 15KI. hIL-15KI mice allow modeling of TEV-53408 that binds specifically to human IL-15, but does not bind to murine IL- 15. hIL- 15 rodents do not respond to the YTE mutation that prolongs antibody half-life in humans. Rodents also generate an antidrug antibody (ADA) immune response against the human TEV-53408 that limits the amount of time that TEV-53408 effects can be studied.Dose-response relationship between TEV-53408 and IL-15

[0381] Male IL15KI mice (n=40) were divided into 2 groups and administered a single dose of 0.05 or 1 mg / kg (n=3 / time point / group) of antibody via subcutaneous (sc) administration. Blood samples were collected at predetermined times: pre-dose, Ih, 4h, 24h, 72h, 120h and 168h post-treatment per group. Additional samples were received from the mouse NK cell study below, where male and female IL15KI mice received a single dose of TEV-53408 at 0.15 and 1 mg / kg via sc (n=3 / time point / group). Blood samples were collected at 72, 168, 240, 336, 504, 672 and 840 h post-treatment. TEV-53408 in serum was determined using research grade LC-MS / MS analytical method. PK parameters were estimated by noncompartmental methods using validated software.

[0382] TEV-53408 systemic exposure was demonstrated in hIL15KI mouse serum following a single sc administration of 0.05, 0.15, and 1 mg / kg. For 0.05 and 1 mg / kg, maximum observed drug concentration (Cmax) was found to be dose proportional. Dose normalized Cmax values demonstrated linear PK. However, the dose normalized area under the drug concentration time curve from time 0 to infinity (AUCo- / ) values were more than dose proportional, which could be due to animal variability or target-mediated drug disposition (TMDD). The terminal half-life (t%) ranged from 2 to 4 days.Table 53: Composite Mean Pharmacokinetics Parameters of TEV-53408 in 8 Week Old hIL15KI Mice Administered a Single sc Doses of 0.05, 0.15 and 1 mg / kg1Data should be considered with caution since full profile was not captured2Outlier animals were excluded from the analysisAUC=area under the serum concentration-time curve; AUCo-28d=AUC from time 0 to 28 days after dosing; AUCo-t=AUC from time 0 to the time of the last measurable drug concentration; AUCo- / = AUC from time 0 to infinity; Cmax=maximum observed serum drug concentration; tmax= time to maximum observed concentration; sc=subcutaneous

[0383] IL-15 levels were highly correlated with TEV-53408 levels, and there was an inverse relationship between free and total IL- 15 levels (Figures 31A-31B). TEV-53408 at 0.15 mg / kg (Figure 31B) or at 1 mg / kg (Figure 31A) resulted in reduction of free IL-15 levels to below level of quantification (BLQ) 3 days post-dose. Free IL- 15 levels returned at 2-weeks timepoint, surpassing the baseline levels to demonstrate a slight rebound effect at 3-weeks timepoint. Total IL- 15 levels (containing both free and TEV-53408-bound IL- 15) showed a mirror picture, decreasing sharply from 2-weeks to 3-weeks timepoints. This is consistent with TEV- 53408 binding all available free IL- 15 initially and more as it is generated until a peak is reached. The week after the peak in total IL-15, the free IL-15 levels rise rapidly and the total goes down. In mice, this can also reflect metabolism of TEV-53408, which is much faster than occurs in humans.NK cell, but not T-cells are decreased by antibody in a dose-responsive manner

[0384] hIL-15KI mice were dosed with a single sc administration of TEV-53408 at 0.017, 0.05, 0.15, 1 mg / kg and sacrificed 7 days following treatment. Blood and spleen samples were obtained to determine treatment effect on both T and NK cells. Flow cytometry was performed on single cells stained with fluorescently-labeled target antibodies (anti-CD3, anti NK1.1).

[0385] Flow cytometry analysis showed no difference between treated and untreated mice in T cell percentages. 0.15 and 1 mg / kg doses significantly decreased NK cells in blood (p<0.01) and spleen (p<0.005). Lower doses did not reach statistical significance but did follow antibody dose.TEV-53408 affects NK cell maturation, but does not affect NK precursors

[0386] hIL-15KI mice were administered a single sc injection of 0.15 mg / kg of anti-IL15 TEV-53408 or KLH-reactive IgG4 isotype control, and sacrificed 2 or 4 weeks following treatment, to assess the time-dependent effect on NK cells developmental stages, as well as the reversibility of the effect. Flow cytometry analysis of the peripheral blood and spleen showed reduction in percentages of NK cells 2 weeks after treatment. Markers included CDl lb, CD127, NK 1.1, Lin (CD3, CD 19, Ly6C, Ly6G), CD 122, CD27, and CD49b. A full recovery in both peripheral blood and spleen NK cells was observed 4 weeks after TEV-53408 administration, and NK cell levels reached the levels of the isotype control group.Table 54: Flow cytometry summary results of NK cell percentages after 1, 2 and 4 weeks following TEV-53408 administration in spleen and blood of hIL-15KI mice.

[0387] The analysis of the mice bone marrow 2 weeks after treatment demonstrated a reduction in stages B, C, D, and E two weeks post treatment. These stages represent the immature and mature NK cells. No changes were observed in the NK progenitor stage. Similar to blood and spleen, immature and mature NK cells recover in the bone marrow 4 weeks post treatment (Table 54).Table 55: Flow cytometry summary results of NK developmental stages 2 and 4 weeks following TEV-53408 administration in the bone marrow of hIL-15KI mice.

[0388] In conclusion, the bone marrow analysis conducted in mice demonstrated that the effect of TEV-53408 is limited to immature and mature NK cells, is reversible, and has no effect on NK progenitor cells.Summary

[0389] The behavior of TEV-53408 in hIL15 KI mice reflects the observations made in a human clinical trial, monkeys, and the HUMIMIC bone marrow on a chip. TEV-53408 showed a dose dependent reduction in free serum IL- 15 to BLQ. The antibody reduced blood and spleen NK cells, but not T-cells. NK maturation in mouse bone marrow was reversibly inhibited. \Example 13: TEV-53408 demonstrates efficacy in animal model of vitiligo

[0390] A vitiligo model was established in hIL-15KI mice by inducing an immune response to tyrosinase-related protein 2 (TRP2) peptide, a melanocyte-specific enzyme involved in pigment color. (You S, et al Melanocyte-specific CD8+ T cells are associated with epidermal depigmentation in a novel mouse model of vitiligo. Clin Exp Immunol. 2013;174(l):38-44).

[0391] 8 -10 weeks old hIL-15KI female mice were immunized with TRP2 peptide (amino acids 180-188) with the adjuvants LPS and CpG-ODN subcutaneously into the hind paw (2 weekly injections) and then intradermally into the tail dermis (2 weekly injections) for a total of 4 injections over 4 weeks (Figure 32A). About six weeks after the first immunization, depigmentation of the tail skin was observed starting at the injection point, then expanding throughout the tail (Figure 32B). Depigmentation was accompanied by accumulation of TRMS in the skin, as evident by analysis of T cells isolated from the tail skin. TRMS are long-lived resident memory CD8 T cells, defined by the expression of specific surface markers, such as CD103, and are the effector cells in vitiligo, and are the proposed target of TEV-53408. Vitiligo was accompanied by melanocyte loss as shown by loss of TRP2-staining cells. These results show the success of this model in hIL-15 knock in mice, and a mechanistic communality between the human disease and this animal model.

[0392] TEV-53408 was administered once a week as subcutaneous injection at 5 mg / kg for3 consecutive weeks, starting 2 weeks after the last immunization. Saline or IgG4 isotype control antibody were used as negative controls (Figure 32A). Pigment degree and extent was calculated by imaging software. Mice were euthanized 24 hours after the last dose, and spleen, blood and tail skin epidermis were analyzed. Figure 32B shows a comparison of normal pigmentation (top 3images) against vitiligo (bottom three). Clear depigmentation can be seen in the mouse tails, and this is accompanied by a lack of melanocytes and loss of TRP2-staining tissues.

[0393] 3 weekly doses of 5 mg / kg antibody SC TEV-53408 significantly attenuated depigmentation (*P<0.05: unpaired t-test, parametric, two tails) (Figure 33 A). Skin tissue was obtained, CD8 T-cells extracted, and FACS analysis performed, gating for live —> immune cells (CD45+) — T cells (CD45+CD3+) — CD8+ — CD 103. TRP2 immunization increased CD8 TRM cells in the skin compared to naive and this was significantly reduced by TEV-53408 (**P<0.01, Figure 33B). To calculate the number and type of activated TRM cells, extracted CD8 T-cells were exposed to TRP2, then examined for ZFNy and TNFa secreting clones with FACS. Naive mice, and cells exposed to DMSO did not exhibit increased ZFNy or TNFa. (Figure 33C). TRP2 did not activate CD8 T-cells from naive mice, but did activate CD8 T-cells previously immunized with TRP2 and exhibiting vitiligo. This is consistent with a resident population of TRP2-reactive cells in vitiligo skin that would inhibit the reestablishment of melanocytes.

[0394] Administration of TEV-53408 significantly (p<0.05) decreased ZFNy response to TRP2. TEV-53408 reduced TNFa secreting cells, but this difference was non-significant (p=0.13). Given that n=6 in each group, it is possible that a larger sample size would clarify this result. In a different cohort of mice, TNFa+ cells were significantly inhibited by TEV-53408 at week 8.5. At week 8.5 timepoint, ZFNy is not significant, so this could alternatively be a kinetic issue, where suggesting specific kinetics of the immune response where ZFNy rises first and then declines as TNFa levels rise. These data show that TEV-53408 inhibits T-cell reactivity in the skin, consistent with less depigmentation.

[0395] TEV-53408 also significantly reduced CD8 memory T-cells in the circulation, compared to saline (Gating: live immune cells (CD45+) T cells (CD45+CD3+) CD8+ CD127+KLRG1- Statistics: unpaired t-test, parametric, two-tailedp<0.01, Figure 34A). This is relevant to the long term treatment of vitiligo, as central memory T-cell (TCM) can migrate to the skin and become resident (TRM).

[0396] Immunization with TRP2 caused serum IL-15 to rise to 436 pg / ml 4 weeks later in mice administered saline. Those administered TEV-53408 had an average of 28.7pg / ml, a statistically significant difference (p<0.0001, unpaired two-tailed t-test). IL-15 was 329.2pg / ml in naive mice.

[0397] These data confirm the role of TEV-53408 in neutralizing IL- 15 and corresponding treatment of vitiligo. It also shows that, despite the clear relevance of IL- 15 in vitiligo, a rise inIL- 15 by a non-statistically significant 32% over background can accompany disease. Accordingly, IL-15 may be involved in more diseases than currently appreciated, based solely on epidemiologic studies. The model was repeated with different doses of TEV-53408: 0.2 mg / kg, 1 mg / kg, and 5 mg / kg (n= 10). TRP2 immunization induced vitiligo and raised total CD8 (Figure 34B) and CD8 TRM cells (Figure 34C) in the skin. In this model, 1 mg / kg of TEV-53408 was the minimum effective dose to significantly lower total CD8 cells and CD8 TRM cells.

[0398] This model reflects the human clinical situation where resident CD8 TRM act as a continued barrier to melanocyte repopulation of vitiligo-affected skin. These data demonstrate that TEV-53408 can mitigate vitiligo in mice, reducing the numbers of TRP2 reactive skin localized CD8 TRM and CD8 T-cells and circulating resident memory cells, via suppression of IL-15 levels. By suppressing the melanocyte-destroying immune response, the TEV-53408 permits repopulation of affected skin with melanocytes. Because the anti-IL15 TEV-53408 is based on human IgG4 backbone, it is likely that mice develop an immune response against the TEV-53408 that reduces exposure between the 3rdand 4thdose of TEV-53408. This not only requires more frequent dosing that would be necessary in humans, it prevents an extended model to experimentally confirm melanocyte repopulation.Example 14: Comparison of TEV-53408 with other clinical stage IL-15-binding antibodies in healthy hIL15 mice

[0399] Two other IL-15 binding antibodies are in clinical trials, AMG714 (from Amgen, W02004076620) and CALY-002 (WO 2024 / 028448 originally from Calypso Biotech BV, now part of Novartis). The published primary sequence of AMG714 and CALY-002 was used to generate clones, express and purify antibodies. These antibodies are denoted COMP1 (from AMG714) and COMP2 (from CALY-002) to distinguish these antibodies from the originals with which they share the primary amino acid sequence. In clinical trials, AMG714 and TEV-53408 was administered subcutaneously and CALY-002 was administered intravenously. In animal models, therefore, we administered COMP1 and TEV-53408 subcutaneously and COMP2 is administered intraperitoneally. The normal IV route for mice is via the tail, but in the vitiligo model the immunizations are done in the tail and, more importantly the phenotype is seen in the tail so i.v. isn’t practical in this model and risks biasing the results by administration close to the site of the phenotype being studied.

[0400] A single dose of each antibody (sc for TEV-53408 and COMP1, ip for COMP2) was administered at 4 different doses (0.04, 0.2, 1 and 5 mg / kg) into hIL-15KI C57BL / 6 mice.Serum pK was measured at 24 hours and 7 days via cheek puncture. After 7 days, mice were euthanized and NK cells and T-cells were measured in spleen, and free IL-15 and antibody measured in the serum.

[0401] Free IL-15 levels were measured in the serum, demonstrating a dose response for TEV-53408 and for C0MP1. Notably, TEV-53408 was more potent than COMP1 in inhibiting free IL- 15, reaching BLQ levels at 5 mg / kg. COMP2, however, showed a floor effect after only -50% reduction in IL-15 levels, and did not show further reduction (Fig. 35A). This result appears to reflect in vitro studies, where TEV-53408 is more potent than COMP1, and COMP2 inhibited at most 75% compared to the almost 100% of TEV-53408.

[0402] NK cells were strongly diminished by TEV-53408 even in the lower dose (0.04), and were kept to a minimum in the other dose levels. COMP1 also reduced NK cell levels, although less potently than TEV-53408, and both reached -96% reduction in NK cell levels, but TEV-53408 reached this level at the 0.2 mg / kg dose whereas COMP1 reached it at the 5 mg / kg dose. By contrast, and similarly to the effect it had on IL- 15 levels, COMP2 showed a floor effect in reducing NK cell levels, reaching a maximum of 75% reduction in NK cell levels (Fig. 35B).

[0403] The antibodies had no effect on general T cells, as well as specifically on CD8+ T cells in the spleen (Figs. 35C-35D).

[0404] Table 56 shows the data in greater detail:Table 56: data from comparative pK / pD study of TEV-53408 and other IL-15 antibodies

[0405] In conclusion, TEV-53408 was more potent per dose than both COMP1 and COMP2 in reducing IL-15 serum levels and spleen NK cell levels in healthy hIL15 mice.Example 15: Comparison of TEV-53408 with other clinical stage IL-15-binding antibodies in a mouse model of vitiligoVitiligo model establishment

[0406] hIL-15KI mouse was used to establish a vitiligo mouse model, according to You et al. Clin Exp Immunol. 2013 Oct;174(l):38-44. Briefly, mice were immunized with TRP2iso-i88 peptide in the presence of immunostimulants (LPS and CpG ODN). 4 weekly immunizations were performed: the first two were injected intradermally into the footpad, and the last two were injected intradermally into the proximal area of the tail. In this model, wound is formed in the tail injection area, about one week after the last immunization. This wound is resolved and following this resolution, the tail area becomes depigmented. Tail depigmentation spreads from immunization area towards the tail base.TEV-53408 potency vs. other antibodies competitors in the vitiligo mouse model

[0407] This vitiligo mouse model was used for comparing the potency of TEV-53408 to COMP1 and COMP2. The depigmentation (and repigmentation) phenotype is relatively slow and, in immunocompetent mice, cannot be remedied by multiple doses of human antibody over time because of the antidrug antibody responses in mice. Therefore, we also tracked the more sensitive and faster responding cellular markers that are believed to underly the mechanism of vitiligo.

[0408] Two weeks after the last TRP2 immunization, antibodies were administered at 1 mg / kg, for 3 weekly administrations. TEV-53408 and COMP1 were administered subcutaneously, and COMP2 was administered intraperitoneally. The experiment was terminated 7 weeks and 1 day after the beginning of immunizations, 24 hours after the last antibody dosing (Fig. 36A).

[0409] The therapeutic effect of TEV-53408 on tail skin depigmentation signal in this model is usually weak because antidrug antibodies are generated at about 2 weeks post dosing, and the de- and re-pigmentation responses are slow. Hence, we do not reliably reach statistical significance compared to untreated diseased mice if small numbers are tested. In this study, where n=10, tail skin depigmentation was not significantly reduced by TEV-53408 compared to saline group, although a trend towards reduction could be observed. The comparator antibodies also did not reduce depigmentation but, on the contrary, showed a non-significant trend increaseddepigmentation compared to Saline group (Fig. 36B). This implies lower efficacy of comparators compared to TEV-53408 in reducing depigmentation, although firm conclusions cannot be drawn from a statistically non-significant small study.

[0410] As to the more responsive and sensitive cellular markers, skin tissue resident memory CD8+ T cells (Trm) and short-lived CD8+ T effector cells were significantly suppressed by TEV-53408, but not by antibodies COMP1 and COMP2 (Figs. 36C-36D). TEV-53408 was also the only antibody to reduce Trm progenitors in the blood, which may be an important reservoir of the vitiligo immune response (Fig 36E). All three antibodies also suppressed short-lived CD8+ T effector cells in the blood, with COMP2 showing a milder effect (Fig. 36F).

[0411] All antibodies reduced also serum free IL- 15 levels, but to a different degree. In agreement with the T cell results, TEV-53408 was the most potent, reducing levels of IL-15 to BLQ in most of the mice. COMP1 reduced IL-15 but to a lesser extent, and COMP2 reduced IL- 15 only to -50% of Saline group levels (Fig. 2G).

[0412] These data demonstrate superior efficacy of TEV-53408 over competitors in suppressing vitiligo T cell response.No antibody had a significant effect on NK cells in the skin (Fig 37A) but all had significant effects on NK cells in the blood (Fig. 37B). This differentiates the effect on NK cells from T-cell subsets.Example 16. TEV-53408 is effective in a mouse model of alopecia areata

[0413] Alopecia areata (AA) is an autoimmune disease affecting -2% of the population worldwide, in which T cells attack the hair follicle causing hair loss. AA is believed to be triggered by loss of immune privilege of the hair follicle environment. IL- 15 and IL-15Ra were shown to be overexpressed in T cell-invaded hair follicles of Alopecia patients (Xing L et al. Nat Med. 2014 Sep;20(9): 1043-9) and the serum levels of IL-15 correlated with disease severity (Ebrahim AA, et al. Int J Trichology . 2019 Jan-Feb; 11 (l):26-30). Antibodies to rodent IL-2 receptor P (the common IL-2 and IL-15 receptor) prevented alopecia in rodents but was unable to reverse established alopecia, (Xing, Nat Med. 2014; Seok J, et al. Nat Immunol. 2023 Aug;24(8): 1308-1317).

[0414] We employed a humanized mouse model for alopecia described by Gilhar A et al. J Invest Dermatol. 2013 Mar;133(3):844-847. This model used human scalp skin containing hair follicles transplanted onto the back of Beige-SCID immunodeficient mouse. Following innervation and vascularization of the scalp graft, ex -vivo expanded autologous human NKG2D+ CD8+ T cells are injected intradermally into the graft. The injected NKG2D+ CD8+ T cells attack the hair follicle, causing Alopecia (Fig. 38A).

[0415] We evaluated the efficacy of TEV-53408 in this model using two regimens: (1) a preventative regimen, in which TEV-53408 administration started together with T cell injection, and (2) a therapeutic regimen, in which TEV-53408 administration started after alopecia was established. TEV-53408 was administered sc twice-a-week at 5 mg / kg. The JAK inhibitor Tofacitinib (5 mg / kg, po, 5 days-a-week) was used as a positive control, and PBS and anti-KLH IgG4 (5 mg / kg, sc, twice-a-week) were used as negative controls. Because the mice are immunocompromised we are able to inject multiple doses of antibodies without inducing anti-drug antibodies.

[0416] TEV-53408 prevented alopecia in mice treated with the preventative regimen, almost to the extent of Tofacitinib (Fig. 38B **p<0.01)).

[0417] TEV-53408 also reversed alopecia in mice treated with the therapeutic regimen(Fig. 38C, *p<0.05). This result distinguishes TEV-53408 from antibodies to rodent IL-2 receptor P that were unable to reverse established alopecia (Xing, Nat Med. 2014; Seok J, et al. Nat Immunol. 2023 Aug;24(8): 1308-1317). TEV-53408 also shows that inhibition of IL-15 alone, without also inhibiting IL-2 or other molecules that bind to IL-2RP, can prevent and treat alopecia.

[0418] Tofacitinib outperformed TEV-53408 in this model. It should be noted, however, that tofacitinib has an FDA “black box warning” regarding an increased risk of serious heart- related events, cancer, blood clots, and death. JAKi in general also have elevated risks of serious infections such as pneumonias and tuberculosis. A further differentiator is the JAKi are known for a fast recurrence of alopecia upon treatment withdrawal. TEV-53408 has longer half-life which can make its effect last longer, plus less side effects will omit the need to stop treatment

[0419] Histology studies are performed on thin sections of the skin, passing through the hair bulb, to show the effect of TEV-53408 on the immune response in the skin generally, and the hair bulb in particular.

[0420] A follow-up study in mice is performed with different doses of TEV-53408, to show a PK / PD correlation, and with COMP1 and COMP2 comparators.Example 17. TEV-53408 is effective in a mouse model of Atopic Dermatitis

[0421] Atopic Dermatitis (AD) is a chronic inflammation of the skin, manifested as dry, itchy and inflamed skin. The incidence of AD is -12% in children and -5% in adults. AD is mainly driven by a type 2 helper T cell (Th2) response, but eosinophils and CD8+ T cells are also involved. IL- 15 is increased in the skin of AD patients (Karlen H, et al Int Arch Allergy Immunol. 2020; 181(6):417-421) and was shown to regulate Th2 cytokines (Mori A. J Immunol. 1996 Apr1 ; 156(7):2400-5). WO 2024 / 028448 reports that lOmg / kg CALY-002 was therapeutic in a mouse model of atopic dermatitis.

[0422] We employed the humanized mouse model of Keren A, et al. Allergy. 2023 Jun;78(6): 1538-1553. Briefly, human abdominal skin is transplanted onto the back of Beige-SCID immunodeficient mouse. Following innervation and vascularization of the skin graft, ex-vivo differentiated autologous human Th2 cells are injected intradermally into the graft, attack the skin and cause AD (Fig. 39A).

[0423] To evaluate the efficacy of TEV-53408 in this model, TEV-53408 was administered sc starting at the time of Th2 cell injection and AD induction. TEV-53408 was administered 3 times-a-week at 2.5 mg / kg. The anti-IL-4R Dupilumab (25 mg / kg, sc), an FDA approved drug for AD, was used as a positive control, and anti-KLH IgG4 (25 mg / kg, sc) or PBS were used as negative controls.

[0424] Representative pictures and histological evaluation of the skin grafts showed that while PBS- and isotype control -treated mice mostly showed atopic skin grafts, TEV-53408-treated mice showed either healthy or only partially atopic skin grafts, and Dupilumab-treated mice all harbored healthy skin grafts (Figs. 39B-39C). Specifically, TEV-53408 reduced some of the AD phenotypes seen in this mouse model, including a statistically significant reduction in erythema, similarly to Dupilumab and near significant reduction in lichenification (Figs. 39D, 39E). Other AD phenotypes such as crusting was not significantly reduced by neither TEV-53408 nor Dupilumab (Figs. 39F).

[0425] Immunohistochemical evaluation of the skin sections were performed on n=8 mice / group (originating from 5 human donors). Filaggrin, Krtl6, TARC and TSLP were quantified using Image J as percentage of staining coverage. Proliferation (Ki67): % of positive cells. For CD8, CD4, IL-4, IL-31, and IL-13: positive cells were counted in an area of 0.66 mm2.

[0426] The data showed that TEV-53408 reduced hyperproliferation and expansion of epidermal cells (as evident by Ki67 and Krtl6 staining), thus leading to reduction in epidermal thickness, a hallmark of AD (Figs. 39G, H, I). TEV-53408 also restored filaggrin expression at the epidermis, which is a pivotal skin barrier protein (Fig 39K).

[0427] Disruption of the epithelial barrier of the skin is a prominent phenotype of AD. Transepidermal water loss (TEWL) is a measure for the amount of water that escapes from the stratum corneum per area of skin, and it reflects skin water barrier integrity. TEV-53408 reduced TEWL measurement, similarly to Dupilumab (Fig 39J). This result aligns with the restoration of filaggrin expression.

[0428] Additional immunohistochemical evaluation of the skin sections for relevant immune cell populations and cytokines related to the Th2 response and AD phenotype, showed that TEV-53408 counteracted the inflammatory reaction incited in this AD mouse model. Upon TEV-53408 administration, the levels of CD4+ and CD8+ T cells were reduced (Figs 40A, 40B), the Th2 cytokines IL-4 and IL- 13 were suppressed (Figs 40C, 40D), and AD-related secreted factors IL-31, TSLP and TARC were inhibited (Figs 40E, 40F, 40G)

[0429] Efficacy with TEV-53408 was observed at 2.5mg / kg, a dose lower than the lOmg / kg of CALY-002 tested in WO 2024 / 028448, or the lOmg / kg of dupilumab, an established AD treatment.Example 18. TEV-53408 Resolved IL-15-Mediated Intestinal Inflammatory Pathology in Transgenic Mice That Overexpress Human IL-15 in Intestinal Epithelial Cells (hIL-15 Tg)

[0430] T3b-hIL-15 transgenic mice express human IL- 15 under a gut epithelial cellspecific promoter (T3b). These mice develop severe duodenojejunal inflammation, with accumulation of NK-like CD8+lymphocytes in the affected mucosa and peripheral lymphoid tissues, and recapitulate many of the defining pathologic features of celiac disease. Two studies were conducted using these transgenic mice. The mice were treated weekly for 3 weeks (4 doses) by intraperitoneal injection with TEV-53408 or control antibodies at doses of 0.3 or 3 mg / kg in the first study and 0.1, 0.3, or 1 mg / kg in the second study. At study endpoint (day 28), blood and spleen samples were prepared for immunophenotyping by flow cytometry, and the small intestine was removed and prepared for histopathologic evaluation. Total splenocyte counts and splenic cytotoxic T cell (CD3+CD8+) and NK cell (CD3 NK1.U) counts were significantly decreased in animals treated with TEV-53408 at 3 mg / kg or 1 mg / kg when compared to the control IgG4-treated group. Blood cytotoxic T cells were significantly reduced at 1 or 3 mg / kg, while NK cells were reduced at 0.3, 1, and 3 mg / kg but not at 0.1 mg / kg when compared to the control IgG4-treated group. Hematoxylin and eosin staining was used to examine small intestine tissue histopathology, and immunohistochemistry was performed to identify CD3+lymphocytes present in the small intestine of treated animals. TEV-53408 dosed at 1 mg / kg or 3 mg / kg resolved the IL-15-mediated mononuclear cell inflammation, villous atrophy, and crypt hyperplasia observed in control animals.Example 19. TEV-53408 pharmacokinetics / pharmacodynamics in monkeys

[0431] The pharmacokinetics (PK) of TEV-53408 was thoroughly characterized in a pharmacologically relevant species with cross-reactivity to TEV-53408, the cynomolgus monkey,in a single-dose PK / pharmacodynamics (PD) study following intravenous (iv) administration. Multiple dose toxicokinetic (TK) data after iv and subcutaneous (sc) administration was obtained in cynomolgus monkeys from the non-Good Laboratory Practice (GLP) Dose-Range Finding (DRF) study and the 13 - and 26 -week GLP toxicology studies. In addition, a single dose PK study was performed in a humanized IL- 15 mice (hIL15KI).

[0432] Following single iv administration at doses of 1 or 10 mg / kg in cynomolgus monkeys, TEV-53408 had low clearance and prolonged serum t%. The mean volume of distribution (Vz) was less than total body water, suggesting that TEV-53408 is confined largely to the circulatory system. Following single and multiple sc administrations (doses up to 200 mg / kg), median tmax was observed 2 to 5 days postdose. Bioavailability following sc administration was >80%. The t / 2was generally similar after iv and sc administrations and ranged from 18 to 25 days. Systemic exposures appeared to increase proportionally to dose over a wide range of doses (1 to 400 mg / kg iv and 50 to 200 mg / kg sc). Repeated sc dosing in monkeys (every 4 weeks [q4w]) resulted in approximately 1.5- to 2. Of old accumulation relative to day 1. No sex-related differences in PK were observed.

[0433] The serum concentrations of the free IL-15 (ie, IL-15:IL15Ra complex not bound by TEV-53408) and the total IL-15 (ie, free and bound to TEV-53408) have been assessed in all in vivo NHP studies. The relationship between TEV-53408 exposure, free IL-15 and total IL-15 concentrations in serum, as well as the blood NK cell counts, showed the following: TEV-53408 binds to IL- 15 efficiently, and concentrations of free IL- 15 in serum decrease below the limit of quantification (BLQ) shortly after dosing and remain BLQ until TEV-53408 concentrations in serum decrease sufficiently. Total IL-15 levels in serum increased until they reached a plateau for all dose levels, routes of administration, and regimens tested. The decrease in free IL- 15 was accompanied by a reduction in peripheral blood NK cell counts in monkeys, and recovery of the NK cells was demonstrated when the concentration of TEV-53408 in serum was estimated to be <0.1 pg / mL.

[0434] Anti-drug antibody (ADA) formation was observed in several animals across the 4 studies in naive cynomolgus monkeys that were analyzed for ADAs. ADAs did not markedly impact exposures, and in general, did not diminish the pharmacologic response as assessed by IL- 15 and NK cell counts. In general, the ADA response in nonhuman primates is not predictive of the ADA response in humans (International Council for Harmonisation of Technical Requirements for Pharmaceuticals for Human Use [ICH] S6[R1]).- I l l -

[0435] In conclusion, the PK profile of TEV-53408 is typical of a half-life extended mAh with relatively slow sc absorption, low clearance, low volume of distribution, and a long elimination half-life.Example 20. Summary of dose experiments

[0436] Different experiments support a range of doses. No serious adverse health effects were observed in any subject at the maximum applied dose: humans at 2mg / kg of body weight (150 mg), mice dosed up to 5 mg / kg, and monkeys up to 400 mg / kg. Effectiveness in a murine model of celiac disease was observed at a dose of Img / kg, and in monkeys at lOmg / kg. These correspond to human doses of 70 mg and 700 mg.Table 57Example 21. Gluten challenge experiment

[0437] Subjects with celiac disease demonstrated by biopsy and anti-gliadin antibodies are placed on at least 12 months of gluten free diet. Subjects are administered a single dose of 150mg TEV-53408 or placebo control. Two weeks later, all subjects are challenged with a daily 3g gluten dose in the form of a cookie, for 14 weeks, then followed up until week 70. Daily patient-reported outcomes are collected. Blood samples are taken at week 1 (prior to antibody administration), week 5 (early gluten challenge), week 13 (late challenge), and later to measure the levels of NK and other immune cells, antibody response, and biomarkers.

[0438] Participants with celiac disease tolerate TEV-53408 and exhibit a decline in NK cell numbers similar to that observed in healthy normal volunteers. TEV-53408-induced repair ofthe gut is demonstrated by serum I-FABP levels. Gluten challenge induces a rise in serum I-FABP in placebo control subjects, demonstrating gluten-induced enteropathy. Subjects previously administered TEV-53408 do not exhibit a gluten induced rise in serum I-FABP. Over time, even in the presence of gluten challenge, TEV-53408-treated subjects exhibit a further decline in serum I-FABP, indicating continued repair of the gut.

[0439] It is to be appreciated that the Detailed Description section, and not the Summary and Abstract sections, is intended to be used to interpret the claims. The Summary and Abstract sections can set forth one or more but not all exemplary aspects of the present invention as contemplated by the inventor(s), and thus, are not intended to limit the present invention and the appended claims in any way.

[0440] The foregoing description of the specific aspects will so fully reveal the general nature of the invention that others can, by applying knowledge within the skill of the art, readily modify and / or adapt for various applications such specific aspects, without undue experimentation, without departing from the general concept of the present invention. Therefore, such adaptations and modifications are intended to be within the meaning and range of equivalents of the disclosed aspects, based on the teaching and guidance presented herein. It is to be understood that the phraseology or terminology herein is for the purpose of description and not of limitation, such that the terminology or phraseology of the present specification is to be interpreted by the skilled artisan in light of the teachings and guidance.

[0441] The breadth and scope of the present invention should not be limited by any of the above-described exemplary aspects, but should be defined only in accordance with the following claims and their equivalents.

[0442] Various publications, including patents, published applications, accession numbers, technical articles and scholarly articles are cited throughout the specification. Each of these cited publications is incorporated by reference, in its entirety and for all purposes, in this document.Sequence Table:

Claims

WHAT IS CLAIMED IS:

1. An aqueous pharmaceutical formulation, comprising:(a) about 100 mg / mL to about 150 mg / mL of an antibody that specifically binds to human IL-15, wherein the antibody or antigen-binding fragment thereof comprises: a heavy chain variable region complementarity determining region (CDR)1 comprising the amino acid sequence of SEQ ID NO: 1, a heavy chain variable region CDR2 comprising the amino acid sequence of SEQ ID NO: 2, a heavy chain variable region CDR3 comprising the amino acid sequence of SEQ ID NO: 3, a light chain variable region CDR1 comprising the amino acid sequence of SEQ ID NO: 4, a light chain variable region CDR2 comprising the amino acid sequence of SEQ ID NO: 5, and a light chain variable region CDR3 comprising the amino acid sequence of SEQ ID NO: 6.; and(b) one or more pharmaceutically acceptable excipients.

2. The aqueous pharmaceutical formulation of claim 1, wherein the antibody comprises a heavy chain variable region comprising the amino acid sequence of SEQ ID NO: 7 and a light chain variable region comprising the amino acid sequence of SEQ ID NO: 8.

3. The aqueous pharmaceutical formulation of claim 2, wherein the antibody comprises a heavy chain comprising the amino acid sequence of SEQ ID NO: 9 and a light chain comprising the amino acid sequence of SEQ ID NO: 10.

4. The aqueous pharmaceutical formulation of any one of claims 1-3, comprising about 100 mg / mL of the antibody.

5. The aqueous pharmaceutical formulation of any one of claims 1-3, comprising about 150 mg / mL of the antibody.

6. The aqueous pharmaceutical formulation of claim 4, wherein said pharmaceutically acceptable excipients include histidine, sucrose, EDTA and polysorbate.

7. The aqueous pharmaceutical formulation of claim 6, comprising (b) about 10 mM to about 50 mM Histidine; (c) about 150 mM to about 350 mM Sucrose; (d) about 0.01 mM to about 0.5 mM EDTA; and (e) about 0.1 mg / mL to about 0.9 mg / mL Polysorbate-80.

8. The aqueous pharmaceutical formulation of claim 7, comprising about 20 mM Histidine.

9. The aqueous pharmaceutical formulation of claim 7, comprising about 250 mM Sucrose.

10. The aqueous pharmaceutical formulation of claim 7, comprising about 0.1 mM EDTA.

11. The aqueous pharmaceutical formulation of claim 7, comprising about 0.5 mg / mL Polysorbate-80.

12. The aqueous pharmaceutical formulation of any one of claims 1-11, further comprising about 75 mM to about 180 mM Arginine-Hydrochloride (Arg-HCl).

13. The aqueous pharmaceutical formulation of claim 7, comprising about 100 mg / mL of the antibody, about 20 mM Histidine, about 250 mM Sucrose, about 0.1 mM of EDTA, and about 0.5 mg / mL Polysorbate-80.

14. The aqueous pharmaceutical formulation of claim 7 having a pH of 5.2 ± 0.5.

15. The aqueous pharmaceutical formulation of claim 7 having an osmolality of from about 200 mOsm / kg to about 350 mOsm / kg at room temperature.

16. The aqueous pharmaceutical formulation of any one of claims 1-15, having no significant difference in stability after storage at 2-8° C for up to 24 months, 25°C for up to 18 months, or 40°C for up to 6 months.

17. The aqueous pharmaceutical formulation of claim 16 having: a. at least about 98% antibody monomer content after storage at 25 °C for up to 3 months; b. having no significant difference in viscosity, hydrodynamic radius, or polydispersity after storage at 2-8 °C for up to 24 months; c. having no significant difference in sub-visible particle content after storage at 2-8°C for up to 24 months, 25°C for up to 18 months, or 40°C for up to 6 months; d. having from about 60% to about 140% relative potency measured by a cell-based potency assay after storage at room temperature for 24 hours, and after storage at 40°C for up to 1 month;e. having no significant difference in thermal stability after storage at 2-8°C for up to 24 months or 25°C for up to 6 months; f. having no significant difference in secondary and / or tertiary protein structure after storage at 2-8°C for up to 24 months or 25°C for up to 18 months; and / or g. having no significant difference in secondary protein structure after storage at 2-8°C for up to 24 months.

18. A container comprising the aqueous pharmaceutical formulation of any one of claims 1-17.

19. The container of claim 18, wherein the container is selected from(a) a glass vial.(b) a pre-filled syringe.(c) s a polyolefin or polyvinyl chloride bag; and(d) an autoinjector.

20. A method of treating an immune disease in a subject in need thereof, comprising administering to the subject an antibody comprising HCDR 1-3 comprising the amino acid sequences of SEQ ID NOs: 1-3, respectively, and LCDR 1-3 comprising the amino acid sequences of SEQ ID NOs: 4-6, respectively, at a dose from about 10 mg to about 700 mg.

21. The method of claim 20, wherein dose is from about 10 mg to about 50 mg every 4 weeks or about 50 mg to about 150 mg every 12 weeks.

22. The method of claim 21, wherein the dose is selected from a. about 10 mg every 4 weeks; b. about 20 mg every 4 weeks; c. about 30 mg every 4 weeks; d. about 50 mg every 4 weeks; e. about 50 mg every 12 weeks, and f. about 150 mg every 12 weeks.

23. The method of claim 20, wherein the immune disease is an autoimmune disease.

24. The method of claim 23, wherein the autoimmune disease is selected from celiac disease, vitiligo, alopecia, atopic dermatitis, eosinophilic esophagitis, Sjogrens syndrome, graft- versus host disease, type I diabetes, and rheumatoid arthritis.

25. The method of claim 23, wherein the treatment of autoimmune disease is characterized by (a) a reduction in the number of activated CD8 T resident memory cells and effector cells at a site of disease, (b) reduction in circulating CD8 T memory cells (c) reduction in inflammatory cytokines and / or (d) reduction in pathology.

26. The method of claim 25 wherein the reduction in inflammatory cytokines is reduction in any one of IFNy, TNFa, IL-4, IL- 13, IL-31, TSLP and / or TARC27. The method of claim 25 wherein the reduction in effector cells is reduction in CD4+ and / or CD8+ T cells.

28. The method of claim 20, wherein the method of treating an immune disease in a subject in need thereof comprising administering to the subject does not cause serious adverse effects.

29. The method of claim 28, wherein the administering to the subject does not cause a decline in circulating NK cell numbers below 1% of total lymphocytes; does not significantly affect the CD56brightand %CD56dimratio in circulating NK cells and / or does not significantly affect NK cell function in remaining NK cells.

30. The method of claim 28, wherein NK cell function in remaining NK cells is assessed by CD 107a expression or perforin expression.

31. The method of claim 28, wherein administering to the subject inhibits NK cell maturation to stage 4 in the bone marrow, but does not affect NK cell progenitors.

32. The method of claim 20 for the treatment of celiac disease, wherein the treatment of celiac disease in the subject comprises: (a) repairing the mucosa of a small intestine; (b) increasing the mean villous height vs. crypt depths (V / C); and / or (c) decreasing symptoms of one or more of muscle pain, body pain, joint pain, fatigue, bloating, gas, nausea, cramps, constipation, diarrhea, skin rash, headache, migraine headache, depression, anxiety, brain fog, and irritability; as assessed by patient reported outcomes.

33. The method of claim 20 for the treatment of vitiligo, wherein the treatment of vitiligo in the subject comprises (a) reduction in the rate of skin depigmentation, (b) increase in the area of skin pigmentation (c) reduction in the number and / or activity of melanocytetargeting T-cells in the skin and / or (d) reduction in the number and / or activity of melanocyte-targeting T-cells in other organs.

34. The method of claim 33, wherein skin pigmentation is measured by F-VASI and / or T- VASI.

35. The method of claim 33, wherein the subject is further treated with ultraviolet light.

36. The method of claim 20 for the treatment of alopecia areata, wherein the treatment of alopecia areata in the subject comprises (a) reduction in the rate of hair loss, (b) increase in hair (c) reduction in the number and / or activity of hair follicle-targeting T-cells in the skin and / or (d) reduction in the number and / or activity of melanocyte-targeting T-cells in other organs.

37. The method of claim 20 for the treatment of alopecia areata, wherein the treatment of alopecia areata in the subject is measured by reduction in the level of at least one of the following markers in skin: CD4+T cells, CD8+NKG2D+T cells, and fFNy.

38. The method of claim 20 for the treatment of atopic dermatitis, wherein the treatment of atopic dermatitis comprises (a) reduction of erythema (b) reduction in lichenification (c) reduction in skin thickening (d) reduction in hyperproliferation and expansion of epidermal cells (e) increased expression of filaggrin (f) reduction in transepithelial water loss (TEWL) (g) reduction the levels of CD4+ and CD8+ T cells in the skin and / or (h) suppression of IL- 4, IL-13, IL-31, TSLP and / or TARC.

Images