MEANS AND METHODS FOR PRODUCING ANTIBODY-LINKER CONJUGATES

Abstract

The present invention relates to a method for generating an antibody-payload conjugate by means of a microbial transglutaminase (MTG). The method comprises a step of conjugating a linker comprising the structure (shown in N->C direction) (Sp.sub.1)-RK-(Sp.sub.2)-B-(Sp.sub.3) or (Sp.sub.1)-B-(Sp.sub.2)-RK-(Sp.sub.3) to a Gln residue comprised in an antibody, wherein (Sp.sub.1) is a chemical spacer or is absent; (Sp.sub.2) is a chemical spacer or is absent; (Sp.sub.3) is a chemical spacer or is absent; R is arginine or an arginine derivative or an arginine mimetic; K is lysine or a lysine derivative or a lysine mimetic; B is a linking moiety or a payload; and wherein the linker is conjugated to the Gin residue comprised in the antibody via a primary amine comprised in the side chain of the lysine residue, the lysine derivative or the lysine mimetic. Further, the invention relates to antibody-linker conjugates, antibody-drug conjugates and linker constructs comprising an RK motif.

Claims

1. A method for producing an antibody-linker conjugate by means of a microbial transglutaminase (MTG), the method comprising a step of conjugating a linker comprising the structure (shown in N->C direction)
(Sp.sub.1)-RK-(Sp.sub.2)-B-(Sp.sub.3) or (Sp.sub.1)-B-(Sp.sub.2)-RK-(Sp.sub.3) to a Gln residue comprised in an antibody, wherein (Sp.sub.1) is a chemical spacer or is absent; (Sp.sub.2) is a chemical spacer or is absent; (Sp.sub.3) is a chemical spacer or is absent; R is arginine or an arginine derivative or an arginine mimetic; K is lysine or a lysine derivative or a lysine mimetic; B is a linking moiety or a payload; and wherein the linker is conjugated to the Gln residue comprised in the antibody via a primary amine comprised in the side chain of the lysine residue, the lysine derivative or the lysine mimetic.

2-33. (canceled)

34. An antibody-linker conjugate comprising: a) an antibody; and b) a linker comprising the structure:
(Sp.sub.1)-RK-(Sp.sub.2)-B-(Sp.sub.3) or
(Sp.sub.1)-B-(Sp.sub.2)-RK-(Sp.sub.3); wherein (Sp.sub.1) is a chemical spacer or is absent; (Sp.sub.2) is a chemical spacer or is absent; (Sp.sub.3) is a chemical spacer or is absent; R is arginine or an arginine derivative or an arginine mimetic; K is lysine or a lysine derivative or a lysine mimetic; B is a linking moiety or a payload; wherein the linker is conjugated to the antibody via an isopeptide bond formed between a γ-carboxamide group of a glutamine residue comprised in the antibody and a primary amine comprised in the side chain of the lysine residue, the lysine derivative or the lysine mimetic comprised in the RK motif comprised in the linker.

35-39. (canceled)

40. The antibody-linker conjugate according to claim 34, wherein the linker comprises an amino acid sequence selected from the group consisting of: RKAA (SEQ ID NO:1), RKA (SEQ ID NO:2), ARK (SEQ ID NO:3) and RK-Val-Cit (SEQ ID NO:54).

41-48. (canceled)

49. The antibody-linker conjugate according to claim 43, wherein B is a toxin selected from the group consisting of a pyrrolobenzodiazepine; an auristatin; a maytansinoid; a duocarmycin; a nicotinamide phosphoribosyltransferase (NAMPT) inhibitor; a tubulysin; an enediyne; an anthracycline derivative (PNU); a pyrrole-based kinesin spindle protein (KSP) inhibitor; a cryptophycin; a drug efflux pump inhibitor; a sandramycin; an amanitin; and a camptothecin.

50. (canceled)

51. (canceled)

52. The antibody-linker conjugate according to claim 43, wherein B is a payload, and the chemical spacer (Sp.sub.2) comprises a p-aminobenzyl carbamoyl (PABC) moiety.

53-59. (canceled)

60. The antibody-linker conjugate according to claim 34, wherein the antibody is selected from the group consisting of: Brentuximab, Trastuzumab, Gemtuzumab, Inotuzumab, Avelumab, Cetuximab, Rituximab, Daratumumab, Pertuzumab, Vedolizumab, Ocrelizumab, Tocilizumab, Ustekinumab, Golimumab, Obinutuzumab, Sacituzumab, Belantamab, Polatuzumab and Enfortumab.

61. (canceled)

62. (canceled)

63. An antibody-drug conjugate comprising: a) an IgG antibody; and b) a linker comprising a drug moiety B, wherein the drug moiety B is covalently linked to an amino acid sequence selected from the group consisting of: RKAA (SEQ ID NO:1), RKA (SEQ ID NO:2), ARK (SEQ ID NO:3), RKR (SEQ ID NO:4) or RK-Val-Cit (SEQ ID NO:54); wherein the linker is conjugated to the IgG antibody via an isopeptide bond formed between the γ-carboxamide group of glutamine residue Q295 (EU numbering) of the C.sub.H2 domain of the antibody and the primary amine comprised in the side chain of the lysine residue comprised in the linker.

64. (canceled)

65. The antibody-drug conjugate according to claim 63, wherein the drug moiety B is linked to the N- or C-terminus of the amino acid sequence comprised in the linker via a self-immolative moiety, and wherein the self-immolative moiety comprises a p-aminobenzyl carbamoyl (PABC) moiety.

66. The antibody-drug conjugate according to claim 63, wherein the IgG antibody is a glycosylated IgG antibody, in particular wherein the IgG antibody is glycosylated at residue N297 (EU numbering) of the C.sub.H2 domain.

67. The antibody-drug conjugate according to claim 63, wherein the IgG antibody is an IgG1 antibody.

68. The antibody-drug conjugate according to claim 63, wherein the IgG antibody is Polatuzumab or an antibody comprising a heavy chain as set forth in SEQ ID NO:5 and a light chain as set forth in SEQ ID NO:6.

69. The antibody-drug conjugate according to claim 63, wherein the IgG antibody is Trastuzumab or an antibody comprising a heavy chain as set forth in SEQ ID NO:7 and a light chain as set forth in SEQ ID NO:8.

70. The antibody-drug conjugate according to claim 63, wherein the IgG antibody is Enfortumab or an antibody comprising a heavy chain as set forth in SEQ ID NO:9 and a light chain as set forth in SEQ ID NO:10 or 11.

71. The antibody-drug conjugate according to claim 63, wherein the drug is a toxin selected from the group consisting of: a pyrrolobenzodiazepine; an auristatin; a maytansinoid; a duocarmycin; a nicotinamide phosphoribosyltransferase (NAMPT) inhibitor; a tubulysin; an enediyne; an anthracycline derivative (PNU); a pyrrole-based kinesin spindle protein (KSP) inhibitor; a cryptophycin; a drug efflux pump inhibitor; a sandramycin; an amanitin; and a camptothecin.

72. The antibody-drug conjugate according to claim 63, wherein the linker has the structure RKAA-PABC-B.

73. The antibody-drug conjugate according to claim 63, wherein the linker has the structure RKA-PABC-B.

74. The antibody-drug conjugate according to claim 63, wherein the linker has the structure ARK-PABC-B.

75. The antibody-drug conjugate according to claim 63, wherein the linker has the structure RKR-PABC-B.

76. The antibody-drug conjugate according to claim 63, wherein the linker has the structure RK-Val-Cit-PABC-B.

77. A linker construct comprising the structure:
(Sp.sub.1)-RK-(Sp.sub.2)-B-(Sp.sub.3) or
(Sp)-B-(Sp.sub.2)-RK-(Sp.sub.3); wherein (Sp.sub.1) is a chemical spacer or is absent; (Sp.sub.2) is a chemical spacer or is absent; (Sp.sub.3) is a chemical spacer or is absent; R is arginine or an arginine derivative or an arginine mimetic; K is lysine or a lysine derivative or a lysine mimetic; B is a linking moiety or a payload.

78-85. (canceled)

86. The linker construct according to claim 77, wherein: the linker construct comprises the structure RKAA-B, RKA-B, or ARK-B.

87-95. (canceled)

96. The linker construct according to claim 77, wherein the linker construct comprises the structure RKAA-PABC-B, RKA-PABC-B, ARK-PABC-B, B-PABC-RKR, or RK-Val-Cit-PABC-B; wherein B is an auristatin or a maytansinoid.

97-103. (canceled)

104. A pharmaceutical composition comprising the antibody-linker conjugate according to claim 34 and at least one pharmaceutically acceptable ingredient.

105. (canceled)

106. (canceled)

107. A method of treating a neoplastic disease, a neurological disease, an autoimmune disease, an inflammatory disease or an infectious disease, said method comprising administering to a patient in need thereof the antibody-drug conjugate according to claim 63.

108. The method according to claim 107, wherein: the antibody-drug conjugate comprises Polatuzumab and the neoplastic disease is a B-cell associated cancer; the antibody-drug conjugate comprises Trastuzumab and the neoplastic disease is a HER2-positive cancer; or the antibody-drug conjugate comprises Enfortumab or an Enfortumab variant and the neoplastic disease is a Nectin-4 positive cancer.

109-116. (canceled)

117. A pharmaceutical composition comprising the antibody-drug conjugate according to claim 63 and at least one pharmaceutically acceptable ingredient.

118. A method of treating a neoplastic disease, a neurological disease, an autoimmune disease, an inflammatory disease or an infectious disease, said method comprising administering to a patient in need thereof the antibody-linker conjugate according to claim 34.

119. The antibody-drug conjugate according to claim 74, wherein B is an auristatin or a maytansinoid.

120. The antibody-drug conjugate according to claim 119, wherein the auristatin is MMAE and wherein the maytansinoid is DM1 or maytansine.

121. The antibody-drug conjugate according to claim 63, wherein the linker has the following structure: ##STR00008##

Description

BRIEF DESCRIPTION OF THE FIGURES

[1200] FIG. 1 shows the chemical structure of RKAA-MMAE linker-payload complex according to this invention, wherein the N-terminally protected Ac-RKAA peptide is covalently linked to PABC and MMAE.

[1201] FIG. 2 shows the chemical structure of RKA-MMAE linker-payload complex according to this invention, wherein the N-terminally protected Ac-RKA peptide is covalently linked to PABC and MMAE.

[1202] FIG. 3 shows the chemical structure of ARK-MMAE linker-payload complex according to this invention, wherein the N-terminally protected Ac-ARK peptide is covalently linked to PABC and MMAE.

[1203] FIG. 4 shows the chemical structure of KRA-MMAE linker-payload complex (not according to this invention), wherein the N-terminally protected Ac-KRA peptide is covalently linked to PABC and MMAE.

[1204] FIG. 5 shows the chemical structure of AKR-MMAE linker-payload complex (not according to this invention), wherein the N-terminally protected Ac-AKR peptide is covalently linked to PABC and MMAE.

[1205] FIG. 6 shows the chemical structure of KAAR-MMAE linker-payload complex (not according to this invention), wherein the N-terminally protected Ac-KAAR peptide is covalently linked to PABC and MMAE.

[1206] FIG. 7 shows the chemical structure of KARA-MMAE linker-payload complex (not according to this invention), wherein the N-terminally protected Ac-KARA peptide is covalently linked to PABC and MMAE.

[1207] FIG. 8 shows the chemical structure of RKAA-maytansine linker-payload complex according to this invention, wherein the N-terminally protected Ac-RKAA peptide is covalently linked to PABC and maytansine.

[1208] FIG. 9 shows the chemical structure of maytansine-RKR linker-payload complex according to this invention, wherein maytansine is covalently linked to a C4-alkyl-amide spacer connected to RKR peptide (having a C-terminal amide protecting group).

[1209] FIG. 10 shows the size exclusion chromatogram (SEC) of the antibody-drug conjugate ARA-01-RKAA-PABC-MMAE of the invention.

[1210] FIG. 11 depicts the results of the dose-dependent in vitro cytotoxic effects of the antibody-drug conjugate ARA-01-RKAA-PABC-MMAE of the invention against three different CD79b overexpressing cell lines (a-c) and a CD79b non-expressing cell line (d).

[1211] FIG. 12 shows the plasma concentrations of polatuzumab (SEQ ID NOs: 5 and 6), antibody-drug conjugate of the invention (ARA01-RKAA-PABC-MMAE) and polatuzumab-vedotin (Polivy®) at different time-points after a single i.v. injection of 5 mg/kg into CD1 Swiss mice. The concentration in plasma was determined by ELISA. Mean plasma concentrations of 5 mice are plotted versus time, error bars represent standard error of mean (SEM). It is to be understood that the abbreviation ARA01-RKAA-MMAE in FIG. 12 refers to the antibody-drug conjugate ARA01-RKAA-PABC-MMAE.

[1212] FIG. 13 shows schematically the one-step conjugation process.

[1213] FIG. 14 shows the chemical structure of ARK-PEG2-PABC-MMAE linker-payload complex according to this invention, wherein the N-terminally protected Ac-ARK peptide is covalently linked to a PEG2 spacer and PABC-MMAE payload.

[1214] FIG. 15 shows the chemical structure of ARK-PEG2-(NH)—(CH.sub.3)—S—C4-maytansine linker-payload complex according to this invention, wherein the N-terminally protected Ac-ARK peptide is covalently linked to a PEG2-(NH)—(CH.sub.3)—S—C4 alkyl spacer and the maytansine payload.

[1215] FIGS. 16A-16B depict the Granta 519 mouse tumor model. Human B-cell lymphoma-tumor cells (Granta 519) were inoculated subcutaneously in CB17 SCID mice (n=8 per treatment group). When tumors reached a size of about 200 mm.sup.3, the animals received a single injection of 0.53 mg/kg or 2.1 mg/kg polatuzumab-vedotin (Polivy®) and either 0.53 mg/kg, 1 mg/kg or 2.1 mg/kg of ARA01-RKAA-PABC-MMAE. ARA01-RKAA-PABC-MMAE provided equal tumor growth inhibition and survival at about half the payload dose relative to polatuzumab-vedotin (FIG. 16A, and 0.53 mg/kg doses in FIG. 16B). At an approximately equal payload dose relative to polatuzumab-vedotin, ARA01-RKAA-PABC-MMAE treatment led to a greater antitumor efficacy and a considerable survival advantage with 6/8 complete tumor remissions over polatuzumab-vedotin with 0/8 complete tumor remission (comparison of the 0.53 mg/kg dose of polatuzumab-vedotin and 1 mg/kg dose of ARA01-RKAA-PABC-MMAE in FIG. 16B). Mean tumor volumes are shown ±standard error of the mean (SEM).

[1216] FIG. 17 shows the chemical structure of an RKAA-payload complex according to this invention, wherein the N-terminally protected Ac-RKAA peptide is covalently linked to the exatecan derivative Dxd via a self-immolative methyl amine linker.

[1217] FIG. 18 shows the chemical structure of an RKAA-payload complex according to this invention, wherein the N-terminally protected Ac-RKAA peptide is covalently linked to the payload PNU-159682 via the self-immolative moiety PABC-EDA.

[1218] FIG. 19 shows the chemical structure of an RKAA-payload complex according to this invention, wherein the N-terminally protected Ac-RKAA peptide is covalently linked to the payload Duocarmycin GA via the self-immolative moiety PABC-EDA.

[1219] FIG. 20 shows the chemical structure of an RKAA-payload complex according to this invention, wherein the N-terminally protected Ac-RKAA peptide is covalently linked to the payload PBD via the self-immolative moiety PABE.

[1220] FIG. 21 shows the chemical structure of an RKAA-payload complex according to this invention, wherein the N-terminally protected Ac-RKAA peptide is covalently linked to the payload maytansine via a self-immolative methyl amine linker and an additional alkyl linker molecule.

[1221] FIG. 22 shows the chemical structure of an RKR-payload complex according to this invention, wherein the N-terminally protected Ac-RKR peptide is directly coupled to the payload MMAE.

[1222] FIG. 23 shows the chemical structure of an RKAA-payload complex according to this invention, wherein the N-terminally protected Ac-RKAA peptide is covalently linked to the payload Duocarmycin GA via a self-immolative para-methyl aniline (PMA) moiety.

[1223] FIG. 24 shows the chemical structure of an RKR-payload complex according to this invention, wherein the C-terminally protected RKR-amide peptide is covalently linked to the payload PNU-159682 via a carbamate.

[1224] FIG. 25 shows the chemical structure of an RKR-payload complex according to this invention, wherein the C-terminally protected RKR-amide peptide is covalently linked to the payload PNU-159682 via a the self-immolative moiety OHPAS-PHB-EDA and an additional (PEG).sub.2 moiety.

[1225] FIG. 26 shows the chemical structure of an RKR-payload complex according to this invention, wherein the C-terminally protected RKR-amide peptide is covalently linked to the payload PBD via a the self-immolative moiety OHPAS and an additional (PEG).sub.2 moiety.

[1226] FIG. 27 shows the chemical structure of an RKR-payload complex according to this invention, wherein the C-terminally protected RKR-amide peptide is covalently linked to the payload PBD via a the self-immolative moiety OHPAS.

[1227] FIG. 28 shows the chemical structure of an RKR-payload complex according to this invention, wherein the C-terminally protected RKR-amide peptide is directly coupled to the payload DM21.

[1228] FIG. 29 shows the chemical structure of an RKR-payload complex according to this invention, wherein the C-terminally protected RKR-amide peptide is covalently linked to the payload DM4 via an alkyl linker molecule comprising a carboxyl group and a thiol group.

[1229] FIG. 30 shows the chemical structure of an RKR-payload complex according to this invention, wherein the C-terminally protected RKR-amide peptide is covalently linked to the payload MMAE via a dicarboxylic acid linker molecule.

[1230] FIG. 31 shows the chemical structure of an RKR-payload complex according to this invention, wherein the C-terminally protected RKR-amide peptide is covalently linked to the payload MMAE via a the self-immolative moiety OHPAS-PHB and an additional (PEG).sub.2 moiety.

[1231] FIG. 32 shows the chemical structure of an RKR-payload complex according to this invention, wherein the C-terminally protected RKR-amide peptide is covalently linked to the payload MMAE via a the self-immolative moiety OHPAS-PHB.

[1232] FIG. 33 shows the chemical structure of an RKR-payload complex according to this invention, wherein the C-terminally protected RKR-amide peptide is covalently linked to the payload Duocarmycin GA via a the self-immolative moiety OHPAS-quaternary ammonium and an additional (PEG).sub.2 moiety.

[1233] FIG. 34 shows the chemical structure of an RKR-payload complex according to this invention, wherein the C-terminally protected RKR-amide peptide is covalently linked to the payload Duocarmycin GA via a the self-immolative moiety OHPAS-quaternary ammonium.

[1234] FIG. 35 depicts the results of the dose-dependent in vitro cytotoxic effects of the antibody-drug conjugate Trastuzumab-RKAA-PABC-MMAE or Trastuzumab-RKAA-PABC-Maytansine of the invention against the HER2-positive cell line SKBR-3.

[1235] FIG. 36 depicts the results of the dose-dependent in vitro cytotoxic effects of the anti-CD79b antibody-drug conjugates ARA01-ARK-PABC-MMAE, ARA01-RKA-PABC-MMAE and ARA01-RKValCit-PABC-MMAE of the invention against the CD79b-positive cell line Granta-519.

[1236] FIG. 37 depicts the results of the dose-dependent in vitro cytotoxic effects of the anti-Nectin-4 antibody-drug conjugates ARA04-RKAA-PABC-MMAE, ARA04-ARK-PABC-MMAE, ARA04-RKA-PABC-MMAE and ARA04-RKValCit-PABC-MMAE of the invention against the Nectin-4-positive cell line SUM190PT.

[1237] FIG. 38 depicts the results of the dose-dependent in vitro cytotoxic effects of the anti-Nectin-4 antibody-drug conjugates ARA04-RKAA-PABC-MMAE, ARA04-ARK-PABC-MMAE, ARA04-RKA-PABC-MMAE and ARA04-RKValCit-PABC-MMAE of the invention against the Nectin-4-negative cell line A549.

[1238] FIG. 39 shows the plasma concentrations of polatuzumab/ARA01 (SEQ ID NOs: 5 and 6) antibody-drug conjugates of the invention (ARA01-ARK-PABC-MMAE, ARA01-RKA-PABC-MMAE, ARA01-RKValCit-PABC-MMAE,) and polatuzumab-vedotin (Polivy®) at different time-points after a single i.v. injection of 5 mg/kg into CD1 Swiss mice. The ADC concentration in plasma was determined by ELISA. Mean plasma concentrations of 5 mice are plotted versus time, error bars represent standard error of mean (SEM). It is to be understood that the abbreviation ARA01-ARK/RKA/RKValCit-PABC-MMAE in FIG. 39 refers to the antibody-drug conjugates ARA01-ARK/RKA/RKValCit-PABC-MMAE.

[1239] FIG. 40 shows the plasma concentrations of enfortumab/ARA04 (SEQ ID NOs: 9 and 11) antibody-drug conjugates of the invention (ARA04-RKAA-PABC-MMAE, ARA04-ARK-PABC-MMAE, ARA04-RKA-PABC-MMAE, ARA04-RKValCit-PABC-MMAE,) and enfortumab-vedotin (Padcev®9) at different time-points after a single i.v. injection of 5 mg/kg into CD1 Swiss mice. The ADC concentration in plasma was determined by ELISA. Mean plasma concentrations of 5 mice are plotted versus time, error bars represent standard error of mean (SEM). It is to be understood that the abbreviation ARA04-ARK/RKA/RKValCit-PABC-MMAE in FIG. 39 refers to the antibody-drug conjugates ARA04-ARK/RKA/RKValCit-PABC-MMAE.

[1240] FIG. 41 depicts the Ramos mouse tumor model. Human Burkitt's lymphoma-tumor cells (Ramos) were inoculated subcutaneously in CB17 SCID mice (n=6 per treatment group). When tumors reached a size of about 200 mm.sup.3, the animals received a single injection of 1.43 mg/kg polatuzumab-vedotin (Polivy®) or 1.25 mg/kg (payload adjusted dose to polatuzumab vedotin) ARA01-RKAA-PABC-MMAE or ARA01-ARK-PABC-MMAE. Both ADCs of this invention provided equal tumor growth inhibition and survival and long-lasting anti-tumor response at about half the payload dose relative to polatuzumab-vedotin. In contrast, polatuzumab vedotin dosed at the same payload dose, only showed a transient tumor elimination. Mean tumor volumes are shown ±standard error of the mean (SEM).

[1241] FIG. 42 depicts the SUM190PT mouse tumor model. Breast cancer-tumor cells (SUM190PT) were inoculated into the mammary fatpat of CB17 SCID mice (n=6 per treatment group). When tumors reached a size of about 200 mm.sup.3, the animals received a single injection of 1.5 mg/kg enfortumab-vedotin (Padcev®9) or 3 mg/kg (payload adjusted dose to enfortumab vedotin) ARA04-RKAA-PABC-MMAE or ARA04-ARK-PABC-MMAE, respectively. Both ADCs of this invention provided complete and long-lasting anti-tumor responses lasting over 103 days at same payload doses as enfortumab-vedotin. In contrast, enfortumab vedotin dosed at the same payload dose, showed a transient anti-tumor response. When looking at FIGS. 42 and 43 it is evident, that the ADCs of this invention, ARA04-RKAA-PABC-MMAE or ARA04-ARK-PABC-MMAE, showed superior efficacy even when only ¼.sup.th of the payload dose (=4-fold less) was adminstered. The non-binding mAb-RKAA-PABC-MMAE did not show any effect on tumor growth. Mean tumor volumes are shown ±standard error of the mean (SEM).

[1242] FIG. 43 depicts the SUM190PT mouse tumor model. Breast cancer-tumor cells (SUM190PT) were inoculated into the mammary fatpat of CB17 SCID mice (n=6 per treatment group). When tumors reached a size of about 200 mm.sup.3, the animals received a single injection of 0.5 mg/kg enfortumab-vedotin (Padcev®9) or 1 mg/kg (payload adjusted dose to enfortumab vedotin) ARA04-RKAA-PABC-MMAE or ARA04-ARK-PABC-MMAE. Both ADCs of this invention provided complete and long-lasting anti-tumor responses lasting over 103 days at same payload doses as enfortumab-vedotin. In contrast, enfortumab-vedotin dosed at the same payload dose resulted only in a mild tumor-growth retardation. Mean tumor volumes are shown ±standard error of the mean (SEM).

EXAMPLES

Example 1: Conjugation of Peptide-MMAE Linkers to Two Different Antibodies

Methods

[1243] The antibody trastuzumab was commercially available (Herceptin®, Roche, bought from a pharmacy), as well as all linker-payload constructs (custom synthesized by Levena Biopharma). Polatuzumab with heavy and light chain consisting of the sequences of SEQ ID NOs: 5 and 6 were transiently transfected into suspension-adapted CHO-K1 cells and expressed in serum-free/animal component-free media. The proteins were purified from the supernatants by Protein A affinity chromatography (Mab Select Sure column; GE Healthcare).

[1244] For the 1-step conjugation (see FIG. 13), 5 mg/ml of native, glycosylated monoclonal antibody in 50 mM Tris pH 7.6, microbial transglutaminase (MTG, Zedira) at a concentration of 5 U/mg in 50 mM Tris pH 7.6 or water, and 5 molar equivalents of the indicated linker-payload were used and incubated for 24 hours at 37° C. in a rotating thermomixer. Conjugation efficiency was assessed by LC-MS under DTT reduced conditions. Reduction of samples was achieved by incubation of antibody-drug-conjugates (ADCs) for 10 min at 37° C. in in 50 mM DTT/50 mM Tris buffer. Probes were analyzed in a Xevo G2-XS QTOF (Waters) coupled to an Acquity UPLC H-Class System (Waters) and a ACQUITY UPLC BEH C18 Column. Conjugation efficiency was calculated from deconvoluted spectra and presented in % linker-payload conjugated antibody (=ADC). Intensities resulting from both glycoforms (G1F and G0F) were taken into account for the calculation of total conjugation efficiency, i.e.,

Total Conjugation efficiency (%)=total intensity−% intensity of unconjugated antibody,

leading to the following formula:

Conjugation efficiency (%)=100*(1−(intensity(G1F)+intensity(G0F))/total intensity))

Results

[1245] The conjugation efficiency varied depending on the linker structure and antibody used, however, it could be observed that conjugation efficiency was highest when the lysine containing peptide linkers comprised an RK motif (Tables 3 and 4).

TABLE-US-00004 TABLE 3 Conjugation efficiency of linker-payload complexes (according to this invention) are shown Linker-payload Conjugation efficiency (%) to Conjugation efficiency (%) to (according to this invention) antibody trastuzumab antibody polatuzumab RKAA-MMAE (FIG. 1) 100% 89% RKA-MMAE (FIG. 2) 100% 93% ARK-MMAE (FIG. 3) 100% 94%

TABLE-US-00005 TABLE 4 Conjugation efficiency of linker-payload (NOT according to this invention) complexes are shown Conjugation Conjugation Linker-payload efficiency efficiency (NOT according (%) (%) to to to antibody antibody this invention) trastuzumab polatuzumab KRA-MMAE (FIG. 4) 43% 49% (SEQ ID NO: 50) AKR-MMAE (FIG. 5) 68% 27% (SEQ ID NO: 51) KAAR-MMAE (FIG. 6) 64% 28% (SEQ ID NO: 52) KARA-MMAE (FIG. 7) 77% 65% (SEQ ID NO: 53)

Example 2: Conjugation of Peptide-Maytansine to Two Different Antibodies

[1246] In order to demonstrate that high conjugation efficiencies could be achieved also with another payload, ie, other than MMAE, maytansine containing linker-payload constructs were used for conjugation to two different antibodies.

Methods

[1247] The conjugations were performed exactly the same way as described in Example 1. The corresponding maytansine-linker constructs were custom synthesized by Levena Biopharm.

Results

[1248] When using a different payload than MMAE, in this example maytansine, the conjugation efficiency was also very high when the lysine containing peptide linkers comprised an RK motif (table 4). This example shows that irrespective of the payload the conjugation efficiency is high when the lysine containing peptide linkers comprises an RK motif.

TABLE-US-00006 TABLE 5 Conjugation efficiency of linker-payload complexes (according to this invention) are shown Linker-payload Conjugation efficiency (%) to Conjugation efficiency (%) to (according to this invention) antibody trastuzumab antibody polatuzumab RKAA-maytansine (FIG. 8) 92% 82% Maytansine-RKR (FIG. 9) 96% 95%

Example 3: Conjugation of Linker-Payloads (According to this Invention) to a Third Antibody

[1249] In order to further demonstrate the high conjugation efficiency obtained with linker-payload constructs (according to this invention) a third antibody was chosen and successfully conjugated with high efficiency (for two different payloads, further demonstrating the universal applicability).

Methods

[1250] The conjugations were performed exactly the same way as described in Example 1. The antibody Enfortumab with heavy chain consisting of the sequence of SEQ ID NO: 9 and light chain variant consisting of the sequence of SEQ ID NO: 10 were transiently transfected into suspension-adapted CHO-K1 cells and expressed in serum-free/animal component-free media. The proteins were purified from the supernatants by Protein A affinity chromatography (Mab Select Sure column; GE Healthcare).

Results

[1251] High conjugation efficiencies were obtained with the antibody Enfortumab using two different linker-payload constructs according to this invention.

TABLE-US-00007 TABLE 6 Conjugation efficiency of linker-payload complexes (according to this invention) are shown Linker-payload Conjugation efficiency (%) to (according to this invention) antibody Enfortumab RKAA-MMAE (FIG. 1) 91% RKAA-maytansine (FIG. 8) 97%

Example 4: Conjugation of Linker-Payloads (According to this Invention) Comprising Non-Amino Acid Spacers

[1252] In order to further demonstrate the high conjugation efficiency obtained with linker-payload constructs (according to this invention), linkers having polyethelene glycol (PEG) spacers were used and conjugated to two different antibodies with high efficiency.

Methods

[1253] The conjugations were performed exactly the same way as described in Example 1. All linker-payload constructs were custom synthesized by Levena Biopharma.

Results

[1254] High conjugation efficiencies to two different antibodies were obtained with linker-payloads (according to this invention) comprising PEG spacers.

TABLE-US-00008 TABLE 7 Conjugation efficiency of linker-payload complexes (according to this invention) are shown Linker-payload (according to this Conjugation efficiency (%) to Conjugation efficiency (%) to invention) antibody trastuzumab antibody polatuzumab ARK-PEG2-PABC-MMAE 99% 92% (FIG. 14) ARK-PEG2-S-C4- 99% 90% maytansine (FIG. 15)

Example 5: ADCs of the Invention are Monomeric and do not Aggregate

[1255] The linker-payload RKAA-PABC-MMAE (FIG. 1) was conjugated to the antibody Polatuzumab (SEQ ID NOs: 5 and 6) as described in Example 1 above. The resulting ADC termed ARA-01-RKAA-PABC-MMAE had a drug-to-antibody ratio (DAR) of 1.9 (determined using standard mass spectrometry methodologies essentially as described in Richard Y. C. Huang and Guodong Chen (2016) Characterization of antibody-drug conjugates by mass spectrometry: advances and future trends, Drug Discover Today Volume 21, Number 5) was analyzed by size exclusion chromatography.

Methods

[1256] Size exclusion chromatography (SEC) was performed using an ÄKTA FPLC (Amersham Pharmacia Biotech) with a Superdex™ 200 Increase 10/300 (Amersham Pharmacia Biotech) column. Proteins were detected with UV/VIS at a wavelength of 280 nm. Samples were analyzed at a flow rate of 1 mL/min in a 50 mM phosphate, 100 mM NaCl, pH 7.4 running buffer.

Results

[1257] Size exclusion chromatography (SEC) profiles after purification demonstrated that ARA-01-RKAA-MMAE eluted as single, monomeric peak showing that the ADC has excellent biophysical properties (FIG. 10).

Example 6: ADCs of the Invention Show Potent Anti-Tumor Effects In Vitro

Methods

[1258] The growth inhibitory effect of ARA01-RKAA-PABC-MMAE was investigated in vitro on the following three CD79b over-expressing cell lines: Granta-519 (DSMZ, Acc No: 342), BJAB (CLS) and WSU-DLCL2 (DSMZ, ACC 575). As a negative control the CD79 negative cell line HT (ATCC, Ref: CRL-2260) was used. 4000 cells were seeded into 96-well culture plates and incubated with ARA-01-RKAA-PABC-MMAE for 72 hours at 37° C. in a humidified chamber and 5% CO.sub.2.

[1259] The viability of the treated cultures was determined by ATP-quantification in a CellTiterGloLuminescence Assay as described by the supplier (Promega). The % viability relative to untreated cells was calculated according to the formula:

[00001]$\%\mathrm{viability}=\left(\frac{O{D}_{\mathrm{experimental}}-O{D}_{\mathrm{blank}}}{O{D}_{untreated}-O{D}_{\mathrm{blank}}}\right)\times 100$

[1260] The average % viability was plotted against log.sub.10(concentration), and the resulting dose-response curves were analyzed by nonlinear regression with the software Prism8, using a four parameter dose-response curve equation.

Results

[1261] FIG. 11 shows that ARA01-RKAA-PABC-MMAE had a very high cytotoxic activity against CD79b over-expressing cells with EC.sub.50 values comparable to conventional ADCs. The cytotoxic activity was highly selective towards CD79b over-expressing cells, as there was essentially no decrease of cell viability in the HT cell line, as expected. In summary, ARA01-RKAA-PABC-MMAE showed an antigen-specific, significant anti-proliferative activity in vitro.

Example 7: ADCs of the Invention Show Favorable Pharmacokinetic Parameters In Vivo

[1262] The pharmacokinetic profile of the anti-CD79b ADC according to this invention ARA01-RKAA-MMAE was investigated in mice and compared to the commercially available anti-CD79b ADC polatuzumab-vedotin (Polivy®). Polatuzumab-vedotin is an ADC consisting of the anti-CD79b antibody polatuzumab wherein MMAE was conjugated to cysteines of the antibody, leading to an average of 3.5 linked MMAE moieties per antibody (European Medicines Agency, Assessment Report on Polivy®, Procedure number: EMEA/H/C/004879/0000, see https://www.ema.europa.eu/en/medicines/human/EPAR/polivy).

Methods

[1263] ARA01-RKAA-PABC-MMAE (produced in-house as described in Example 5 above), Polivy® (Roche, bought from a pharmacy) and the naked anti-CD79b antibody polatuzumab (SEQ ID NOs: 5 and 6; expressed and purified as described above) were injected intravenously into 5 female mice (CD1 Swiss, Janvier) at a dose of 5 mg/kg of ADC or antibody respectively. After 10 minutes, 5.5, 24, 48, 96, 144, 168 and 360 hours, approximately 20 μl of blood was drawn from the vena saphena into EDTA-coated Microvettes CB 300 (Sarstedt). Blood samples were centrifuged for 10 min at 9500×g and the plasma was stored at −80° until the ELISA analysis was performed. Using dilution series with known concentrations of the corresponding sample, concentration in plasma was determined by ELISA using His-tagged human CD79b as capturing agent: 125 ng of HisCD79b (SinoBiological, Ref.: 29750-H08H) diluted in PBS was added to Nickel plates (Ni-NTA HisSorb, Qiagen) and after blocking with 200 μl PBS, 4% milk (Rapilait, Migros, Switzerland), 50 μl of diluted plasma samples (in PBS, 4% milk) was added. After incubation for 1 h and washing with PBS, either total antibody was detected by addition of donkey-anti-human IgG-HRP (Biolegend, Poly24109) to the wells, or, for total ADC detection, a rabbit anti-MMAE antibody (Levena, Ref: LEV-PAE1) was added for another hour at room temperature, washed and detected via anti-rabbit IgG-HRP. Peroxidase activity was detected by addition of 3,3′,5,5′-Tetramethylbenzidine (Sigma) and stopped by the addition of acid. The readout was measured after 1 to 5 min at 450 nm. From the concentrations of the samples determined by ELISA in plasma at different time points after injection and the resulting slope k of the elimination phase (time-points 24h-360h) (plotted in a semi-logarithmic scale), the half-lives (t.sub.1/2) of the samples were calculated using the formula t.sub.1/2=ln2/−k.

Results

[1264] The plasma concentrations measured in the samples taken at different time points after injection are shown in FIG. 12. The half-lives of the ARA01-RKAA-PABC-MMAE and Polivy® are given below in the table 4. It can be seen that the ADC according to this invention, ARA01-RKAA-PABC-MMAE, had at least a 2 times longer half-life than the approved ADC Polivy® in vivo. The improved stability in plasma may lead to a better safety profile as the payload seems not to be released prematurely.

TABLE-US-00009 TABLE 8 Plasma half-lives Construct Half-life (t.sub.1/2), hours Polatuzumab (SEQ ID NOs: 5 and 6), naked 385 antibody ARA01-RKAA-MMAE, intact ADC 248 polatuzumab-vedotin (Polivy ®), intact ADC 120

Example 8: Anti-CD79b ADC of the Invention Inhibits Tumor Growth In Vivo More Efficiently than the Approved Anti-CD79b ADC Polatuzumab-Vedotin

[1265] The anti-CD79b ADC ARA01-RKAA-PABC-MMAE was investigated in vivo for tumor growth inhibition and was compared to the commercially available polatuzumab-vedotin.

Methods

[1266] 20×10.sup.6 human B-cell lymphoma tumor cells Granta 519(DSMZ, Acc No: 342) were implanted s.c. into CB17 SCID mice (Janvier). Tumor dimensions and body weights were recorded three times weekly. The tumor volume was calculated according to the formula volume=(width).sup.2×length×0.5. When the average tumor size reached about 200 mm.sup.3, mice were allocated using a non-random stratification protocol into the treatment groups comprising eight mice each. ARA01-RKAA-PABC-MMAE (produced in-house as described in Example 5 above) at doses of 0.53 mg/kg, 1 mg/kg and 2.1 mg/kg and polatuzumab vedotin at doses of 0.53 mg/kg and 2.1 mg/kg were administered in a single i.v. injection on day 0 (day of randomization). Mice in the control group were injected with PBS. All mouse experiments were performed in accordance with Swiss guidelines and were approved by the Veterinarian Office of Zürich, Switzerland. Following these guidelines, mice had to be sacrificed at day 10 for the PBS group and all 0.53 mg/kg doses, and 2 mice in the 1 mg/kg group at day 6 and 30 (ulcerations of the tumor).

Results:

[1267] The in vivo efficacy of ARA01-RKAA-PABC-MMAE (DAR 1.9) as compared with polatuzumab-vedotin (DAR 3.5) was assessed against the Granta 519 tumor model. Specifically, the animals received a single injection of 0.53 mg/kg or 2.1 mg/kg polatuzumab-vedotin (Polivy®) and either 0.53 mg/kg, 1 mg/kg or 2.1 mg/kg of ARA01-RKAA-PABC-MMAE. Importantly, ARA01-RKAA-PABC-MMAE provided equal tumor growth inhibition and survival at about half the payload dose relative to polatuzumab-vedotin (comparison of the 2 mg/kg doses, see FIG. 16A, and 0.53 mg/kg doses in FIG. 16B). At an approximately equal payload dose relative to polatuzumab-vedotin, ARA01-RKAA-PABC-MMAE treatment led to a greater antitumor efficacy and a considerable survival advantage with 6/8 complete tumor remissions over polatuzumab-vedotin with 0/8 complete tumor remission (comparison of the 0.53 mg/kg dose of polatuzumab-vedotin and 1 mg/kg dose of ARA01-RKAA-PABC-MMAE in FIG. 16B). In summary, at approximately equal payload doses, ARA01-RKAA-PABC-MMAE treatment resulted in greater antitumor efficacy and a considerable survival advantage over polatuzumab vedotin.

Example 9: Conjugation of Various RK-Motif-Peptides to Antibody Trastuzumab

[1268] All tested RK-containing peptides conjugated with high efficiency.

Methods

[1269] Conjugation reactions were adapted from conditions described in Example 1. In brief, mixing 5 mg/ml of native, glycosylated Trastuzumab antibody, MTG at a concentration of 1.5 U/mg, and 20 molar equivalents of the indicated peptide-linker containing a RK-motif, in Tris 50 mM pH 7.6 for 24 hours at 37° C. in a rotating thermomixer. Conjugation efficiency was assessed by LCMS as follows: Conjugation efficiency (CE) was calculated from deconvoluted spectra and presented in %. Intensities resulting from both glycoforms (G1F and G0F) were taken into account for the calculation, according to the formula:

[00002]$\mathrm{CE}\%=\frac{.Math.\left({\left(\mathrm{Int}\left(G0F+G1F\right)\right)}_{\mathrm{cj}}\right)}{.Math.{\left(\mathrm{Int}\left(G0F+G1F\right)\right)}_{\mathrm{cj}.\mathrm{ncj}}}\mathrm{with}\mathrm{cj}=\mathrm{conjugated}\mathrm{and}\mathrm{ncj}=\mathrm{non}-\mathrm{conjugated}$

Results

[1270] All tested RK-motif-linkers conjugated with well with efficiencies >50% to native, fully glycosylated Trastuzumab as shown in Table 9.

TABLE-US-00010 TABLE 9 Conjugation efficiency of peptide linkers containing the RK-motif to Trastuzumab RK-motif Conjugation peptide linker efficiency (%) HRKHA (SEQ ID NO: 55) 98% HRKAH (SEQ ID NO: 56) 91% RKAH (SEQ ID NO: 57) 91% RKH (SEQ ID NO: 58) 87% RKAA (SEQ ID NO: 1) 86% RKA (SEQ ID NO: 2) 86% RKHA (SEQ ID NO: 59) 86% RKHH (SEQ ID NO: 60) 85% ARKAH (SEQ ID NO: 61) 82% ARKHA (SEQ ID NO: 62) 82% HRK (SEQ ID NO: 63) 81% RKAAH (SEQ ID NO: 64) 81% ARKHH (SEQ ID NO: 65) 80% RKAAA (SEQ ID NO: 66) 80%

Example 10: Conjugation of RK-Motif-Peptides

Method

[1271] Reaction conditions: 5 mg/ml of native, glycosylated Trastuzumab antibody, MTG at a concentration of 5 U/mg, and 5 molar equivalents of the indicated peptide-linker, in Tris 50 mM pH 7.6 for 24 hours at 37° C. in a rotating thermomixer. Conjugation efficiency was assessed by LCMS as described in Example 9.

Results

[1272] The peptides containing the RK-motif conjugated with significant conjugation efficiency as shown in Table 10.

TABLE-US-00011 TABLE 10 Conjugation efficiency of peptide linkers containing the RK-motif to Trastuzumab RK-motif peptide Conjugation linker efficiency (%) RKAAR (SEQ ID NO: 67)  95% RRKAY (SEQ ID NO: 68) 100% RRK (SEQ ID NO: 69)  99% ARKRA (SEQ ID NO: 70)  98%

Example 11: Conjugation of RK-Motif-Linker-Payloads with MMAE

[1273] In order to show that RK-motif-linker-payloads are also suitable for antibody conjugation in one-step, additional linker-payloads containing the RK-motif, using MMAE as a payload, were used for conjugation to Trastuzumab.

Method

[1274] Conjugation reactions were performed by mixing 5 mg/ml of native, glycosylated Trastuzumab antibody, MTG at a concentration of 5 U/mg, and 5 molar equivalents of the indicated linker-payload, in Tris 50 mM pH 7.6 for 24 hours at 37° C. in a rotating thermomixer. Conjugation efficiency was assessed by LCMS as described in Example 9.

Results

[1275] Surprisingly, excellent conjugation efficiencies (above 85%) were obtained using various RK-motif-linker-payloads containing MMAE for conjugation to native, glycosylated Trastuzumab antibody as shown in Table 1IA. Surprisingly, it was observed that conjugation efficiency was significantly lower when the linker-payloads did NOT comprise an RK motif as shown in Tables 11A and 11B.

TABLE-US-00012 TABLE 11A Conjugation efficiencies of RK-motif linker-payloads containing MMAE to Trastuzumab (according to this invention) Conjugation RK-linker-payload with MMAE efficiency (%) RKAA-PABC-MMAE (SEQ ID NO: 1) 100% RKA-PABC-MMAE (SEQ ID NO: 2) 100% ARK-PABC-MMAE (SEQ ID NO: 3) 100% RKAAR-PABC-MMAE (SEQ ID NO: 67) 99% RRKAY-PABC-MMAE (SEQ ID NO: 68) 100% RRK-PABC-MMAE (SEQ ID NO: 69) 96% ARKRA-PABC-MMAE (SEQ ID NO: 70) 89% RKValCit-PABC-MMAE 91% (SEQ ID NO: 54) ARK-PEG2-PABC-MMAE (SEQ ID NO: 3) 99%

TABLE-US-00013 TABLE 11B Conjugation efficiencies of NON-RK-motif- linker-payloads with MMAE (NOT according to this invention). Non-RK-Linker- Conjugation payload with MMAE efficiency (%) KRA-PABC-MMAE (SEQ ID NO: 50) 43% AKR-PABC-MMAE (SEQ ID NO: 51) 68% KR-PABC-MMAE (SEQ ID NO: 71) 46% KAAR-PABC-MMAE (SEQ ID NO: 52) 64% KARA-PABC-MMAE (SEQ ID NO: 53) 77% KAA-PABC-MMAE (SEQ ID NO: 72) 57%

Example 12: Conjugation of RK-Motif Linker-Payloads Using Alternative Payload Classes

[1276] In order to demonstrate the versatility of the linker technology of this invention, various RK-motif linker-payloads were used for conjugation to Trastuzumab. Payloads were selected from the following payload classes: cytotoxins, steroids (Cortisol=CS) and immunomodulators (i.e. STING agonists) were evaluated.

Method

[1277] Conjugation reactions were performed by mixing 5 mg/ml of native, glycosylated Trastuzumab antibody, MTG at a concentration of 5-10 U/mg, and 5-10 molar equivalents of the indicated linker-payload, in Tris 50 mM pH 7.6 for 24 hours at 37° C. in a rotating thermomixer. Conjugation efficiency was assessed by LCMS as described in Example 9.

Results

[1278] Surprisingly, excellent conjugation efficiencies (above 80%) were obtained using various RK-motif linker versions and payload classes as shown in Table 4. It was also surprising to observe that the payload located at the N-terminal position was very well tolerated (as demonstrated by May-C5-RKR).

TABLE-US-00014 TABLE 12 Conjugation efficiency of linker-payloads containing the RK-motif linkers with 3 various payload classes. Conjugation RK-linker-payload with efficiency diverse toxins and drugs (%) RKAA-PABC-May (SEQ ID NO: 1) 92% RKAA-PABC-Exa (SEQ ID NO: 1) 98% RKAA-PABC-EDA-PNU (SEQ ID NO: 1) 99% RKAA-PABE-Amanitin (SEQ ID NO: 1) 90% RKAA-EDA-Cortisol (SEQ ID NO: 1) 88% RKAA-PABC-EDA-STING (SEQ ID NO: 1) 96% RKAAR-PABC-Exa (SEQ ID NO: 67) 99% RKAAR-EDA-CS (SEQ ID NO: 67) 98% ARK-S-C5-May (SEQ ID NO: 3) 98% ARK-PABC-Exa (SEQ ID NO: 3) 94% ARK-PEG2-S-C5-May (SEQ ID NO: 3) 99% May-C5-RKR (SEQ ID NO: 4) 96% RRK-PABC-Exa (SEQ ID NO: 69) 83% Maytansine; Exa: Exatecan-derivative; STING (stimulator of interferon genes; class of immunestimulators); PNU (anthracycline analog).

Example 13: Conjugation of RK-Motif Linker-Payloads to Three Different Antibodies

[1279] To demonstrate the universal applicability of the reaction, a selection of RK-motif-linker-payloads containing MMAE or Maytansine (May) were conjugated to three different antibodies: Trastuzumab, Polatuzumab and Enfortumab variant (heavy chain SEQ ID NO: 9 and light chain SEQ ID: 11).

Method

[1280] Conjugation reactions were performed by mixing 5 mg/ml of the indicated native, glycosylated antibody, MTG at a concentration of 5-10 U/mg, and 5-10 molar equivalents of the indicated linker-payload, in Tris 50 mM pH 7.6 for 24 hours at 37° C. in a rotating thermomixer. Conjugation efficiency was assessed by LCMS as described in Example 9.

Results

[1281] Surprisingly, high conjugation efficiencies were obtained to all three tested antibodies with all tested RK-motif-MMAE or May linker payloads as shown in Table 13.

TABLE-US-00015 TABLE 13 Conjugation efficiency of linker- payloads to three different antibodies Conjugation Conjugation Conjugation efficiency efficiency efficiency RK-linker- (%) with (%) with (%) with payload Trastuzumab Polatuzumab Enfortumab RKAA-PABC-MMAE 100% 98% 96% (SEQ ID NO: 1) ARK-PABC-MMAE 100% 94% 97% (SEQ ID NO: 3) RKA-PABC-MMAE 100% 93% 95% (SEQ ID NO: 2) RKValCit-  91% 97% 87% PABC-MMAE (SEQ ID NO: 54) RKAA-PABC-May  92% 82% 94% (SEQ ID NO: 1) May-C5-RKR  96% 95% NT (SEQ ID NO: 4) NT: not tested

Example 14: Conjugation of RK-Motif Linker-Payload with Different Reaction Conditions

[1282] To demonstrate that conjugation with RK-linker-payloads tolerate a wide variety of the reaction conditions, linker-payload conjugation to Polatuzumab was performed using a range of reaction conditions with varying parameters.

Method

[1283] As standard condition, the following parameters were used: 5 mg/ml of native, glycosylated polatuzumab antibody, MTG at a concentration of 5 U/mg, and 5 molar equivalents of RKAA-PABC-MMAE, in Tris 50 mM, pH 7.6, for 24 hours at 37° C. in a rotating thermomixer. Conjugation efficiency was assessed by LCMS as described above in Example 9.

[1284] The variable parameters are shown in Table 14.

Results

[1285] The RKAA-PABC-MMAE linker-payload conjugated with very high conjugation efficiency over a very large range of reaction conditions: Conjugation efficiencies of >80% were achieved using antibody concentration between 5 to 17 mg/ml, MTG concentration relative to antibody concentration (U/mg) between 2 and 10 U/mg. Further, high conjugation efficiencies were also obtained with molar concentrations of linker versus antibody (2 to 8 equivalents) as well as a very wide range of pH (from pH 6.0 with conjugation efficiency of 67% and pH 8 of 86%).

[1286] Surprisingly, higher conjugation efficiency was obtained with less linker-payload excess vs antibody, ie, 2-20 equivalent of linker payload yielded higher conjugation efficiencies than using 80 equivalents, which is contrary to what can be expected. (Table 14).

TABLE-US-00016 TABLE 14 Conjugation efficiency of RK-linker-payload to Polatuzumab under different reaction conditions Effects on reaction parameters on conjugation efficiency Antibody final concentration (mg/mL) Parameters 5 6 8 10 15 17 CE (%) 98% 95% 95% 96% 86% 95% MTG loading (U/mg) Parameters 2 3 3.5 4.5 5 6 9 10 CE (%) 94% 92% 95% 97% 98% 100% 96% 88% Molar equivalent of RK-linker-payload vs antibody Parameters 2 2.5 3 4 5 6 8 20 80 CE (%) 70% 81% 80% 92% 98% 94% 93% 71% 50% pH Parameters 6.0 6.5 7.0 7.5 7.6 8.0 CE (%) 67% 86% 91% 90% 98% 86%

Example 15: ADCs Containing RK-Motif-PABC-Payload Linker Payloads are Efficient In Vitro Using Three Different Antibodies

[1287] To demonstrate ADCs according to this invention, ie, generated with RK-motif-MMAE/Maytansine linker payloads lead to efficient release and target-specific toxicity on cancer cell lines, Trastuzumab-, Polatzumab- (ARA01), and Enfortumab- (SEQ ID 9 and SEQ ID 11; ARA04)-based ADCs were tested on target-expressing cells using linkers RKAA-PABC-MMAE, RKAA-PABC-Maytansine, ARK-PABC-MMAE, RKA-PABC-MMAE and RKValCit-MMAE.

Method

[1288] The growth inhibitory effect of Trastuzumab-RKAA-PABC-MMAE and Trastuzumab-RKAA-PABC-Maytansine was investigated on HER-2 positive SKBR-3 (ATCC HTB-30) cells, the inhibitory effects of ARA01-ARK-PABC-MMAE, ARA01-RKA-PABC-MMAE and ARA01-RKValCit-PABC-MMAE were tested on CD79b-positive Granta-519 lymphoma cells and the cytotoxic effects of ARA04-RKAA-PABC-MMAE, ARA04-ARK-PABC-MMAE, ARA04-RKA-PABC-MMAE and ARA04-RKVa1Cit-PABC-MMAE was investigated on Nectin-4 positive breast cancer cells, SUM190PT (BIOIVT, 28068A16284) cells. Target dependency and specificity was tested on Nectin-4-negative lung carcinoma cells A549 (ATCC CCL-185). For all conditions, 4000 cells were seeded into 96-well culture plates and incubated with the respective ADCs for 72 hours at 37° C. in a humidified chamber and 5% CO.sub.2.

Results

[1289] FIG. 35 shows that Trastuzumab-RKAA-PABC-MMAE and Trastuzumab-RKAA-PABC-Maytansine of this invention both exert a very high cytotoxic activity against HER-2 over-expressing cells with EC.sub.50 values comparable to conventional ADCs.

[1290] Of the CD79b and Nectin-4 targeting ADCs, ARA01 and ARA04 ADCs containing different linkers of this invention show very high and target-specific cytotoxic activity on target-positive Granta-519 (FIG. 36) and SUM190PT (FIG. 37) cells, respectively. The EC.sub.50 values are in the range of conventional ADCs. In contrast, the same ADCs do not affect target-negative A549 cells (FIG. 38) and show comparable effects as conventional ADCs.

[1291] In summary, all ADCs using RK-motif linker-payloads according to this invention either conjugated to Trastuzumab, Polatuzumab or Enfortumab as parent antibodies showed target-specific and significant anti-proliferative activity in vitro.

Example 16: ADCs Containing RK-Motif-PABC-MMAE Linker Payloads Show Favorable Pharmacokinetic Properties In Vivo

[1292] To assess the in vivo stability of the RK-motif MMAE ADCs, mouse pharmacokinetic studies were performed using different RK-motif linker payloads conjugated to Polatuzumab and Enfortumab.

[1293] The pharmacokinetic profile of the anti-CD79b ADCs ARA01-ARK-PABC-MMAE, ARA01-RKA-PABC-MMAE and ARA01-RKValCit-PABC-MMAE generated with linker-payloads according to this invention and anti-Nectin-4 ADCs generated with different linkers ARA04-ARK-PABC-MMAE, ARA04-RKA-PABC-MMAE and ARA04-RKValCit-PABC-MMAE were investigated in mice and compared to the commercially available anti-CD79b ADC polatuzumab-vedotin (Polivy®) and commercially available anti-Nectin-4 ADC enfortumab-vedotin (Padcev®).

Method

[1294] The pharmacokinetic study was performed as described in Example 7, with adaptations of sampling timepoints: blood samples were drawn after 10 minutes, 4, 48, 96, 168, 264, 336 and 504 hours from the vena saphena. For anti-CD79b ADC detection, the method described in Example 7 was adapted. Anti-Nectin-4 ADC detection was performed in brief as follows: ADC concentration in plasma was determined by ELISA using His-tagged human Nectin-4 as capturing agent: 125 ng of His-Nectin-4 (SinoBiological, Ref.: 19771-H08H) was diluted in PBS and added to Nickel plates (Ni-NTA HisSorb, Qiagen). After blocking with 200 μl PBS, 4% milk (Rapilait, Migros, Switzerland), 50 μl of diluted plasma sample (in PBS, 4% milk) was added. After incubation for 1 h and washing with PBS, total ADC was detected using a rabbit anti-MMAE antibody (Levena, Ref: LEV-PAE1) that was added for another hour at room temperature, washed and detected via anti-rabbit IgG-HRP. Peroxidase activity was detected by addition of 3,3′,5,5′-Tetramethylbenzidine (Sigma) and stopped by the addition of acid. The readout was measured after 1 to 5 min at 450 nm. Half-lives were calculated using the ADC concentrations of the samples in plasma plotted against time (semi-logarithmic scale). The resulting slope k of the elimination phase using time-points 48-504 h was used to determine the halve life (t.sub.1/2) with the following formula: t.sub.1/2=ln2/−k.

Results

[1295] The plasma concentrations of intact ADCs measured in the samples taken at different time points after injection are shown for anti-CD79b ADCs (FIG. 39) and for Nectin-4 ADCs (FIG. 40).

[1296] The half-lives for ARA01-ARK-PABC-MMAE, ARA01-RKA-PABC-MMAE and ARA01-RKValCit-PABC-MMAE are given below in Table 15. Surprisingly, an approximately 2-fold longer half-life of all ADCs according to this invention was observed compared to the polatuzumab-vedotin was calculated.

[1297] The half-lives for ARA04-RKAA-MMAE, ARA04-ARK-MMAE, ARA04-RKA-MMAE, ARA04-RKValCit-MMAE and Padcev® are given below in the table 16 showing an average 2 to 2.5-fold longer half-lives of these ADCs compared to the the approved enfortumab-vedotin.

[1298] In summary, all ADCs generated according to this invention, using either Polatuzumab or Enfortumab as parent antibody, show 2-2.5-fold improved half-lives as compared to the vedotin benchmarks. This indicates that the ADCs generated with linker-payloads according to this invention result in improved ADC stability in vivo which may lead to an overall better safety profile and therapeutic index (TI) as the payload does not get released prematurely.

TABLE-US-00017 TABLE 15 Plasma half-lives for anti-CD79b ADCs Construct Half-life (t.sub.1/2), hours Polatuzumab/ARA01 (SEQ ID NOs: 5 and 6), naked 393 antibody ARA01-ARK-MMAE, intact ADC 237 ARA01-RKA-MMAE, intact ADC 197 ARA01-RKValCit-MMAE, intact ADC 220 Polatuzumab-vedotin (Polivy ®), intact ADC 118

TABLE-US-00018 TABLE 16 Plasma half-lives for anti-Nectin-4 ADCs Construct Half-life (t.sub.1/2), hours ARA04 (SEQ ID NOs: 9 and 11), naked antibody 303 ARA04-RKAA-MMAE, intact ADC 245 ARA04-ARK-MMAE, intact ADC 242 ARA04-RKA-MMAE, intact ADC 179 ARA04-RKValCit-MMAE, intact ADC 199 Enfortumab-vedotin (Padcev ®), intact ADC 109

Example 17: ADCs Containing RK-Motif-PABC-MMAE Linker Payloads Show More Efficient Tumor Growth Inhibition In Vivo Compared to Benchmarks in CD79b-Positive Liquid and Nectin-4 Positive Solid Tumor Models

[1299] The anti-CD79b ADCs according to this invention, ARA01-RKAA-PABC-MMAE and ARA01-ARK-PABC-MMAE were investigated in vivo for tumor growth inhibition in a Ramos (CD79b-positive, liquid tumor) model. The anti-tumor properties of anti-Nectin-4 ADCs according to this invention ARA04-RKAA-PABC-MMAE and AR04-ARK-PAPBC-MMAE were tested in a SUM190PT (Nectin-4 positive, solid tumor) xenograft model. A non-binding mAb-RKAA-PABC-MMAE control ADC was included to exclude unspecific ADC activity.

Method

[1300] For SUM190PT xenografts, 2×10.sup.6 cells were injected into the mammary fatpad; for Ramos, 20×10.sup.6 cells were injected s.c. into CB17 SCID mice (Janvier). Tumor dimensions and body weights were recorded three times weekly. The tumor volume was calculated according to the formula volume=(width).sup.2×length×0.5. When the average tumor size reached about 200 mm.sup.3, mice were allocated using a non-random stratification protocol into the treatment groups comprising six mice each. ADCs were intraveously injected once on the day of randomization.

[1301] All ADCs were produced in-house as described in Example 5.

[1302] ARA01-RKAA-PABC-MMAE and ARA01-RKAA-PABC-MMAE (both DAR1.9) were injected at a dose of 1.25 mg/kg (corresponding to 25 ug of payload per kg body weight).

[1303] Polatuzumab vedotin (PV, DAR 3.6) was injected at 1.43 mg/kg corresponding to 50 ug/kg payload or double the payload dose of ARA01-ADCs.

[1304] ARA04-RKAA-PABC-MMAE and ARA04-RKAA-PABC-MMAE (both DAR 1.9) were injected at ADC doses of 1 and 3 mg/kg (corresponding to 10 and 30 ug payload per kg body weight) and were compared to enfortumab vedotin (EV, DAR 3.8) at doses of 0.5 mg/kg and 1.5 mg/kg. A non-binding mAb-RKAA-PABC-MMAE ADC (harboring the same linker-payload and DAR as ARA04 ADCs) was injected at 3 mg/kg. Mice in the control group were injected with PBS. All mouse experiments were performed in accordance with Swiss guidelines and were approved by the Veterinarian Office of Zürich, Switzerland.

Results:

[1305] ADCs of this invention ARA01-RKAA-PABC-MMAE and ARA01-ARK-PABC-MMAE were compared to polatuzumab vedotin (PV) in a in a fast-growing Ramos xenograft model. FIG. 41 shows that one i.v. injection of 1.25 mg/kg (or 25 ug payload per kg mouse weight) resulted in highly efficient anti-tumor response in all mice. In contrast, PV dosed at 1.43 mg/kg or 50 ug payload per kg body weight, only showed a short and transient tumor elimination with no tumor-free individuals at 20 days after treatment. In contrast and very surprisingly, ARA01-RKAA-PABC-MMAE and ARA01-ARK-PABC-MMAE showed a durable complete anti-tumor response at half the payload dose.

[1306] In a solid tumor model, ARA04-RKAA-PABC-MMAE and ARA04-ARK-PABC-MMAE were compared to enfortumab-vedotin (EV) in the SUM190PT breast cancer tumor model. Importantly, both ADCs of this invention, ARA04-RKAA-PABC-MMAE and ARA04-ARK-PABC-MMAE were highly efficacious at 1 and 3 mg/kg and resulted in complete tumor eradication and long-lasting response during the whole course of the study (103 days after injection) as shown in FIGS. 42 and 43.

[1307] At approximately equal payload doses relative to enfortumab-vedotin, both, ARA04-RKAA-PABC-MMAE and ARA04-ARK-PABC-MMAE (3 mg/kg) treatments led to a greater and longer lasting antitumor efficacy and a considerable survival advantage with 4/6 complete tumor remissions over EV with 0/6 complete tumor remission (comparison of the 1.5 mg/kg dose of EV and 3 mg/kg dose of ARA04-RKAA-PABC-MMAE in FIG. 42). The non-binding mAb-RKAA-PABC-MMAE did not have any effect on tumor growth and suggest high target-specificity of the targeting ADCs. Strikingly, both ARA04-RKAA-PABC-MMAE and ARA04-ARK-PABC-MMAE ADCs provided equal tumor growth inhibition and survival at about one-third of dose relative to enfortumab-vedotin (comparison of the 1.5 mg/kg=30 ug/kg payload dose) and 1 mg/kg (=10 ug/kg payload dose) ARA04-RKAA/ARK-MMAE doses, see FIGS. 42 and 43. Taken together, at approximately one third of payload doses, ARA04-RKAA-PABC-MMAE and ARA04-ARK-PABC-MMAE treatments resulted in greater and durable antitumor efficacy and a considerable survival advantage over enfortumab vedotin.

[1308] Overall we summarize that anti-CD79b and anti-Nectin-4 ADCs generated with RK-motif linker-payloads according to this invention, consisting of the same antibody and payload as their respective benchmark ADCs, are highly active in vivo and show 2-3-fold superior efficacy providing considerable survival advantage over vedotin-based ADCs, which is highly surprising.