Anti-CLDN1 antibody-drug conjugates comprising mmae and methods of use thereof

Abstract

The disclosure relates to antibody-drug conjugates comprising an anti-Claudin-1 antibody or antigen-binding fragment thereof and an auristatin moiety, connected by a linker, as well as methods of making and using the same.

Description

[0182] First, all anti-CLDNl antibodies were conjugated with pHrodo Deep Red dye and tested in an internalization assay on adherent cells by imaging. As shown in FIG. 15A and 15B, only antibodies HILI, 7A5 and 6F6 were internalized. HILI was the most internalized antibody, 7A5 the least internalized and 6F6 showed an intermediate internalization ability. The total area of pHrodo-positive pixels per cell was quantified at 16.63, 8.89 and 4.29 for HILI, 6F6 and 7A5 respectively (FIG. 15B).

[0183] Then, the internalization of anti-CLDNl antibodies was assessed by FACS on adherent cells. Under these conditions, HILI showed the highest internalization rate with a MFI value at 6h of 40386, while this MFI value was 14388 and 10736 for 6F6 and 7A5, respectively (FIGs. 16A and 16B).

[0184] Altogether, both internalization studies (by imaging and by FACS) demonstrated a much stronger ability of HILI to internalize compared to the other tested anti-CLDNl antibodies. Conclusion and Discussion

[0185] Five commercially available anti-CLDNl antibodies were compared to HILI for their ability to bind to CLDN1 and to undergo a subsequent internalization, since efficient binding and internalization into cancer cells is a desirable characteristic of an antibody that is to be used in an ADC.

[0186] Only three of the tested antibodies were able to bind CLDN1 on non-adherent cells (Ab 7A5, 6F6 and 421203) when CLDN1 extracellular epitopes were fully accessible. But when mimicking a tissue organization with adherent cells, antibody 6F6 showed an intermediate binding and antibody 7A5 showed a very low level of binding to CLDN1-positive cells, while HILI was able to strongly bind to CLDN1.

[0187] The antibodies’ ability to be internalized was also assessed on both adherent and nonadherent cells. In both conditions, HILI was internalized much better than the other two antibodies.

[0188] Altogether, these data demonstrate that not any anti-CLDNl Ab is suitable for use in an ADC approach. Some antibodies can bind to CLDN1 and were not internalized (e.g., antibody 421203). Some antibodies bind to CLDN1, but their binding strongly affected by cell-cell interaction limiting their interest for an ADC (e.g., antibody 7A5). Finally, 6F6 was able to bind to CLDN1 and undergo a CLDN1-dependent internalization, but with a lower efficacy compared to HILI.

[0189] HILI showed good binding and internalization properties to be suitable for an ADC approach. HILI was shown to be superior to all tested antibodies in its ability to be efficiently internalized in NCLH975 overexpressing CLDN1 Example 10: Comparison of Closely Related Antibodies in the Context of ADCs

[0190] This example describes the biological characterization of two variants of anti-human CLDN1 antibodies HILI and H3L3, which have the same CDRs but different VH and VL sequences and different Fc backbones with human IgGl Fc-silent due to the LFLEPS mutation and with a human IgGl wild type sequence, respectively. The two antibodies were compared in an internalization assay in CLDN1-overexpressing NCLH1975 lung cancer cells by imaging. Methods CLND1 Staining at the Cell Membrane

[0191] CLDN1-transduced NCI-H1975 cells were seeded at 25xl03 cells in a flat bottom 96-well plate (collagen-coated PhenoPlate, black, optically clear) in 100 pL of cell culture medium without phenol red and incubated for 18h. Then the medium was removed, cells were washed twice with 200 pL of HBSS. Then, 100 pL of Hoechst solution at 4 pg / mL were added in each well and cells were incubated for 10 min at 37°C. After 3 washing steps with 200 pL of HBSS, 100 pL of pHrodo-labeled HILI, H3L3 or hlgGl control diluted at 10 pg / mL in complete culture medium without phenol red were added and plates were incubated for 30 min at 4°C. to allow antibodies to bind CLDN1 at the cell membrane. Then, the unbound anti-CLDNl antibodies were washed away with 200 pL of HBSS and plates were incubated in an acidic solution of acetic acid (pH 3.5) to allow pH rodo to fluoresce. Images were acquired with Cytation CIO microscope (Agilent) using DAPI and Cy5 channels, 20x objective. Internalization assay by imaging usingpHrodo technology

[0192] 0.5 mg of each antibody were diluted at 2.2 mg / mL in PBS or a similar neutral buffer free of primary amines. 25 pL of a 1 M freshly prepared sodium bicarbonate (pH 8.5) with the kit component C was added to the antibody solution.

[0193] The contents of the 100 pg vial of pHrodo™ Deep Red amine-reactive dye was dissolved in 100 pL of water to prepare a 0.77 mM (1 mg / mL) labeling solution.

[0194] 50 pL of the labeling solution were added to the antibody solution and were incubated for 2 hours at room temperature, protected from light.

[0195] The dye / antibody ratio was determined using the method from AAT Bio website (url:). These ratio were 4.02, 4.66 and 2.48 for HILI, H3L3 and hlgGl control respectively.

[0196] Internalization assay by imaging (protocol without washout step): CLDN1-transduced NCI-H1975 cells were seeded at 25xl03 cells in a flat bottom 96-well plate (collagen-coated PhenoPlate, black, optically clear) in 100 pL of cell culture medium without phenol red and incubated for 18h. Then the medium was removed, cells were washed twice with 200 pL of HBSS. Then, 100 pL of Hoechst solution at 4 pg / mL were added in each well and cells were incubated for 10 min at 37°C. After 3 washing steps with 200 pL of HBSS, 100 pL of pHrodo-labeled HILI, H3L3 or hlgGl control diluted at 10 pg / mL in complete culture medium without phenol red were added and plates were incubated for 2 h, 4 h and 6 h under standard cell culture conditions to allow antibodies to be internalized (pHrodo dye is pH sensitive, little to no fluorescent signal at neutral pH and fluoresces brightly in acidic environments like lysosomal vesicles). When incubation was completed, the medium was removed, and the cells were washed with PBS before a fixation step. Images were acquired with Cytation CIO microscope (Agilent) using DAPI and Cy5 channels, 20x objective.

[0197] Internalization assay by imaging (protocol with a washout step): A similar assay procedure was applied. 100 pL of pHrodo-labeled HILI, H3L3 or hlgGl control diluted at 10 pg / mL in complete culture medium without phenol red were added to the Hoeachst-stained cells and plates were incubated for 30 min at 4°C to allow antibodies to bind CLDN1 at the cell membrane. Then, the unbound anti-CLDNl antibodies were washed away with 200 pL of HBSS and plates were incubated for 2 h, 4 h and 6 h at 37°C to allow antibodies to be internalized. Data analysis

[0198] Data are expressed as mean± S.D. Statistical analyses were calculated by using GraphPad Prism software (GraphPad Software, La Jolla, CA). Results

[0199] HILI and H3L3 anti-CLDNl antibodies were compared for their ability to internalize in CLDN1-positive cells. In both antibodies, the CDR sequences are identical. Only the sequences outside the CDR differ between those antibodies (see Table 1).

[0200] Both HILI and H3L3 antibodies were shown to similarly detect CLDN1 at the cell surface of CLDN1-transduced NCI-H1975 cells (FIG. 17A).

[0201] The ability of HILI and H3L3 to be internalized in CLDN1-transduced NCI-H1975 cells was assessed by two different approaches. pHrodo-conjugated anti-CLDNl antibodies were applied on cells. The kinetics of internalization was determined either without washing step between antibody binding and internalization, or a washing step was performed to remove the excess of unbound antibodies. In both cases, HILI was shown to be internalized more efficiently than H3L3. With the assay procedure without washout step HILI internalization is significantly higher at 6h compared to H3L3 (FIGs. 17B and 17C). With the protocol including a washout step following antibody binding and before internalization kinetics, at all time points (2h, 4h and 6h), HILI was shown to better internalized than H3L3 (FIG. 17D and 17E). Conclusion and Discussion

[0202] Identical CDR sequences are shared between HILI and H3L3 anti-CLDNl antibodies. Despite a similar binding to CLDN1 expressed at the cell surface of NCI-H1975 cells, internalization of HILI was shown to be significantly higher than that of H3L3.

[0203] This feature suggests that HILI is a better candidate for an ADC approach, because of its internalization capabilities.

Claims

1. An antibody-drug conjugate of Formula (I): A-(LJ-L2-L3-D)n(I)or a pharmaceutically acceptable salt or solvate thereof, wherein:A is an anti-Claudin-1 antibody comprising(i) a heavy chain variable region (VH) comprising a heavy chain complementarity determining region (HCDR)l comprising the amino acid sequence of SEQ ID NO: 5, a HCDR2 comprising the amino acid sequence of SEQ ID NO: 6 and a HCDR3 comprising the amino acid sequence of SEQ ID NO: 7; and(ii) a light chain variable region comprising a light chain complementarity determining region (LCDR)l comprising the amino acid sequence of SEQ ID NO: 8, a LCDR2 comprising the amino acid sequence of GAS, and a LCDR3 comprising the amino acid sequence of SEQ ID NO: 10;L1 is a divalent moiety comprising a functional group attached to the anti-Claudin-1 antibody;L2 is a divalent peptide linker comprising at least two amino acids;L3 is a divalent moiety being attached to D via a cleavable bond, such that D is released when the bond is cleaved;D is an auristatin moiety; andn is an integer between about 2 and about 10.

2. The antibody-drug conjugate of claim 1, wherein the anti-Claudin-1 antibody comprises a VH comprising a sequence that is at least 90% identical to the amino acid sequence of SEQ ID NO: 3 or 13.

3. The antibody-drug conjugate of claim 1 or 2, wherein the anti-Claudin-1 antibody comprises a VL comprising a sequence that is at least 90% identical to the amino acid sequence of SEQ ID NO: 4 or 14.

4. The antibody-drug conjugate of any one of claims 1-3, wherein the anti-Claudin-1 antibody comprises a heavy chain comprising the amino acid sequence of SEQ ID NO: 3 and a light chain comprising the amino acid sequence of SEQ ID NO: 4.

5. The antibody-drug conjugate of any one of claims 1-3, wherein the anti-Claudin-1 antibody comprises a heavy chain comprising the amino acid sequence of SEQ ID NO: 13 and a light chain comprising the amino acid sequence of SEQ ID NO: 14.

6. The antibody-drug conjugate of any one of claims 1-5, wherein the anti-Claudin-1 antibody is a monoclonal antibody.

7. The antibody-drug conjugate of any one of claims 1-6, wherein the anti-Claudin-1antibody is a rabbit, mouse, chimeric, humanized or fully human monoclonal antibody.

8. The antibody-drug conjugate of any one of claims 1-7, wherein the anti-Claudin-1 antibody is an IgG isotype.

9. The antibody-drug conjugate of any one of claims 1-8, wherein the anti-Claudin-1 antibody is an IgGl isotype.

10. The antibody-drug conjugate of any one of claims 1-9, wherein -L'-L2-L3- is**s* denotes attachment to D .

11. The antibody-drug conjugate of any one of claims 1-10, wherein D is a monomethyl auristatin E (MMAE) moiety.

12. The antibody-drug conjugate of any one of claims 1-11, wherein D is13. A method of treating cancer in a subject in need thereof, comprising administering to the subject the antibody-drug conjugate of any one of claims 1-12.

14. The method of claim 13, wherein the cancer is urothelial cancer, esophageal cancer, head and neck cancer, lung cancer, colorectal cancer, liver cancer, or breast cancer.

Images