EA046498B1 - PEGYLATED DRUG LINKERS FOR IMPROVED PHARMACOKINETICS OF LIGAND-DRUG CONJUGATES - Google Patents

Description

Cross reference to related applications

This regular application claims priority under 35 USC § 119(e) based on US Serial Nos. 61/891320 filed Oct. 15, 2013, 61/941904 filed Feb. 19, 2014, 61/947742 filed March 4 2014, and 61/975318, filed April 4, 2014, all of which are incorporated herein by reference in their entirety.

List of sequences

The sequence listing designated 2700-00114PC.'^T25.1x1 13 KB, created on October 9, 2014, is incorporated herein by reference.

Prerequisites for creating the invention

There has been great interest in the use of monoclonal antibodies (mAbs) to target cytotoxic agents to cancer cells. When creating antibody-drug conjugates by attaching a cytotoxic agent to an antibody, typically via a linker, various factors must be considered. These factors include the nature and location of the chemical moiety to conjugate the cytotoxic agent, the mechanism of release of the agent, the structural element(s) (if any) that release the cytotoxic agent, and the structural modification of the released free agent, if any. In addition, if the cytotoxic agent is to be released after internalization of the antibody, the structural elements and mechanism of release of the agent must be consistent with the intracellular transport of the conjugate.

Although a number of different drug classes have been evaluated for antibody delivery, only a few drug classes have been shown to be sufficiently active as antibody-drug conjugates while having a suitable toxicity profile to warrant clinical trials. One such class is the auristatins, related to the natural product dolastatin 10. Representative auristatins include MMAE (Nmethylvaline-valine-dolaisoleucine-dolaproline-norephedrine) and MMAF (N-methylvaline-valine-dolaisoleucine-dolaproline-phenylalanine).

MMAE is an example of a cytotoxic agent that is active as a free drug and is highly active when conjugated to a monoclonal antibody (mAb) and released after internalization into cells. MMAE was successfully conjugated to mAb at the N-terminal amino acid of MMAE via a cathepsin B-cleavable peptide linker containing maleimidocaproyl-valine-citrulline (mc-vc-) and a self-cleavable paminobenzyl-carbamoyl (PABC) group, yielding antibody-drug conjugates of the following structure mAb-(mc-vc-PABC-MMAE)p. (In the formula presented, p refers to the number of (mc-vc-PABC-MMAE) units per antibody.) Upon cleavage of the bond between the vc peptide and the self-cleaving PABC group, the PABC group is released from the MMAE, releasing free MMAE.

Another auristatin, MMAF, is relatively less active as a free drug (compared to MMAE) but is highly active when conjugated to an antibody and internalized into cells. MMAF was successfully conjugated to a monoclonal antibody (mAb) at the N-terminal amino acid of MMAF via a cathepsin B-cleavable peptide linker containing maleimidocaproyl-valine-citrulline (mc-vc-) and a self-cleavable paminobenzyl-carbamoyl (PABC) group, yielding antibody- a drug having the structure mAb-(mc-vc-PABC-MMAF) _p , where p refers to the number of (mc-vc-PABC-MMAF) units per antibody. Upon cleavage of the peptide linker, the self-cleaving PABC group is released from MMAF, releasing free MMAF.

MMAF has also been found to be active as a non-cleavable drug-linker conjugate comprising maleimidocaproyl MMAF (mcMMAF). When this conjugate, mAb-(mcMMAF) _p , is internalized into cells, the active substance released is cys-mcMMAF.

Because the linker is non-cleavable, the maleimidocaproyl and cysteine residue of the antibody remains attached to the N-terminus of MMAF. MMAF has also been reported to be active as a C-terminal conjugate linked at the C-terminal amino acid, phenylalanine, to a peptide maleimidocaproyl linker. When this conjugate, (MMAF-peptide-mc) p-mAb is internalized into cells, the active substance, MMAF, is released upon cleavage of the MMAF(phenylalanine) peptide bond.

In animal models, these MMAE and MMAF conjugates demonstrated drug load-dependent reductions in pharmacokinetic properties. In particular, as the number of drug-linker groups attached to each antibody increased, the pharmacokinetic properties of the conjugates decreased.

Therefore, another important factor in the design of conjugates is the amount of drug that can be delivered per targeting agent (i.e., the amount of cytotoxic agents attached to each targeting agent (eg, antibody), referred to as drug loading or content medicinal substance).

- 1 046498

Historically, it has been assumed that higher drug loads are better than lower drug loads (eg, 8 loads versus 4 loads). This was explained by the fact that conjugates having higher loadings would deliver more drugs (cytotoxic agents) to target cells. This rationale was supported by the observations that conjugates with higher drug loadings were more active against cell lines in vitro. Some later studies, however, showed that this assumption was not supported in animal models. Conjugates loaded with drugs, some auristatins, 4 or 8, were observed to have similar activities in mouse models. Hamblett et al., Clinical Cancer Res. 10:7063-70 (2004). Hamblett et al. also reported that overloaded ADCs were more rapidly cleared from the circulation in animal models. This faster clearance suggests a greater impact on pharmacokinetic properties in higher loading agents compared to lower loading agents. Hamblett et al. Additionally, higher loading conjugates had lower MTDs in mice and, as a result, had narrower therapeutic ranges that were reported. Id. In contrast, ADCs with a drug loading of 2 at engineered sites in a monoclonal antibody have been reported to have equal or better pharmacokinetic properties and therapeutic indices compared to some ADCs with a drug loading of 4. For example, see Junutula et al., Clinical Cancer Res. 16:4769 (2010). Thus, recent trends involve the development of ADCs with low drug loadings.

Therefore, there is a need for antibody-drug conjugate formats (and, more broadly, formats for other conjugates) that allow for higher drug loading but will retain the other characteristics of lower loading conjugates, such as suitable pharmacokinetic properties. Surprisingly, the present invention meets these needs.

Brief description of drawings

Fig. 1. Mean tumor volume versus days post-implantation for the L540cy (Hodgkin's Lymphoma) xenograft model at a higher single dose (2 mg/kg) using non-PEGylated ADC formulations, cAC10-[mc-PAB(gluc)MMAE] _p , (cAC10 -1), parallel oriented PEGylated ADC (cAC10-10) and sequentially oriented PEGylated ADC (cAC10-4) with an average drug load of 8 drug groups/Ab.

Fig. 2. Mean tumor volume versus days post-implantation for the Karpas299 xenograft (ALCL) model when administered at a higher single dose (0.6 mg/kg using non-PEGylated ADC (cAC10-1), parallel-targeted PEGylated ADC (cAC1010) and sequential formulations -targeted PEGylated ADC (cAC10-4) with an average drug loading of 8 drug groups/Ab.

Fig. 3. Mean tumor volume versus days post-implantation for the L540cy (Hodgkin's Lymphoma) xenograft model at a lower single dose (0.5 mg/kg) using formulations of non-PEGylated ADC (cAC10-1), parallel oriented PEGylated ADC (cAC10- 10) and a sequentially targeted PEGylated ADC (cAC10-4).

Fig. 4. Mean tumor volume versus days post-implantation for the Karpas299 (ALCL) xenograft model at a lower single dose (0.15 mg/kg) using formulations of a non-PEGylated ADC (cAC10-1), a parallel-targeted PEGylated ADC (cAC1010), and sequence-targeted PEGylated ADC (cAC10-4) with an average drug load of 8 drug groups/Ab.

Fig. 5. Mean tumor volume versus days post-implantation for the Karpas299 xenograft (ALCL) model with a single dose of 0.2 mg/kg using non-PEGylated ADC formulations, cAC10-[MDpr-PAB(gluc)-MMAE] _p (cAC10-14) and a parallel-targeted PEGylated ADC (cAC10-16) with an average drug load of 8 drug groups/Ab (i.e. p=8).

Fig. 6. Mean tumor volume versus days post-implantation for the Ramos (Burkitt's Lymphoma) xenograft model at a single dose of 1 mg/kg using formulations of non-PEGylated ADC, hBUP-IMDpr-PABtgluct-MMAEIp (hBU12-14) and parallel-targeted PEGylated ADC ( hBU12-16) with an average drug load of 8 drug/Ab groups (i.e. p=8).

Fig. 7. Pharmacokinetic profile (total Ab concentration in μg/ml versus time in days) of a rat after a single intravenous 3 mg/kg dose of compositions of the unconjugated cAC10 antibody, its non-PEGylated ADC (cAC10-1), the parallel-oriented PEGylated ADC (cAC1010) and sequentially -targeted PEGylated ADC (cAC10-4) with an average drug load of 8 drug groups/Ab.

Fig. 8. Size exclusion chromatograms for cAC10 ADCs loaded with 8 drug/Ab groups containing non-PEGylated drugs

- 2 046498 va-linkers and parallel-oriented PEGylated drug-linker groups, where the drug-linker group is MDpr-VC-PABA-MMAE, with PEG Groups of different lengths: cAC10-A (non-PEGylated), cAC10-B ( PEG12), cAC10-C (PEG ₂₄ ), cAC10-D (peg ₃₆ ), cAC10-E (PEG ₁₂ + PEG ₃₆ ), cAC10-F (PEG ₂₄ + PEG ₃₆ ) and cAC10-G (PEG ₃₆ + PEG ₃₆ ).

Fig. 9. Pharmacokinetic profile (total Ab concentration in μg/mL versus time in days) of rats following a single intravenous 3 mg/kg dose of cAC10 ADCs loaded with 8 drug/Ab groups containing non-PEGylated drug linkers, where the ADC conjugate is c -AC10-MDpr-VC-PAB-MMAE (cAC10-A) and cAC10-mc-VC-PABA-MMAE (cAC10-K), and parallel-oriented PEGylated drug-linker groups, where the ADC is represented by the structure cAC10-[MDpr (-XD)-PEG ₂₄ ] _P , and -XD is MDpr-VC-PAB-MMAE (cAC10-C) or mc-VC-PABA-MMAE (cAC10-L), and p=8, compared to control cAC10-NAEM conjugates (cAC10-I) having a PEG ₂₄ backbone capped with ethylaminomaleimide (ie, no drug attached).

Fig. 10. Pharmacokinetic profile (total Ab concentration in μg/mL versus time in days) of rats following a single intravenous 3 mg/kg dose of cAC10 ADCs loaded with 8 drug/Ab groups containing non-PEGylated drug-linker groups, where the ADC conjugate is c-AC10-[MDpr-VC-PAB-MMAE]p (cAC10-A), or parallel-oriented PEGylated drug-linker groups, where the ADC is represented by the structure cAC10-[MDpr (-Ch-E)-PEG] _p , where p=8, -XD is MDpr-VC-PAB-MMAE, and PEG is PEG Group of different lengths: PEG12 (cAC10-B), PEG ₂₄ (cAC10-C) and PEG ₃₆ (cAC10-D), each compared to corresponding control conjugates having a PEG scaffold blocked with ethylaminomaleimide (ie, without drug attached): PEG12 (cAC10-H), PEG ₂₄ (cAC10-I) and PEG36 (CAC10-J).

Fig. 11. Tumor volume (mm ² ) versus days after tumor transplantation in the L540cy xenograft model with a single IV dose of 2 mg/kg non-PEGylated ADC: c-AC10-[MDprVC-PAB-MMAE] _p (cAC10-A), versus untreated animals.

Fig. 12. Tumor volume (mm ² ) versus days after tumor transplantation in the L540cy xenograft model with a single intravenous injection of 2 mg/kg parallel-targeted PEGylated ADC (caC10-B): cAC10-[MDpr(-X-D)-PEG ₁₂ ] _p , where p = 8, and -XD represents MDpr-VCPAB-MMAE, compared to an untreated animal.

Fig. 13. Tumor volume (mm ² ) versus days after tumor transplantation in the L540cy xenograft model with a single intravenous injection of 2 mg/kg parallel-targeted PEGylated ADC (cAC10-C): cAC10-[MDpr(-X-D)-PEG ₂₄ ] _p where p = 8 and -XD represents MDpr-VCPAB-MMAE, compared to the untreated animal.

Fig. 14. Mean tumor volume ( ^mm2 ) versus days after tumor transplantation in a breast cancer xenograft model using a non-PEGylated ADC targeting the LIV-1 antigen: hLIV22-[mc-VC-PAB-MMAE)] _p or hLIV22-[MDpr (-XD)-PEG ₂₄ ] _p where p = 8 and -XD is mc-VC-PAB-MMAE, compared to untreated animals.

Fig. 15: Mean tumor volume ( ^mm2 ) versus days after tumor transplantation in the L540cy xenograft model with a single IV dose of 1 or 0.5 mg/kg non-PEGylated ADC: cAC10-[mc-PAB (giuc)-MMAE] _P (cAC10- 1), or its corresponding parallel-oriented PEGylated ADC:cAC10-{mc-[PAB(gluc)-MMAE]-PEG ₂₄ } _p (cAC10-10), where p = 4, compared with untreated animals.

Fig. 16. Mean tumor volume ( ^mm2 ) versus days after tumor transplantation in the Karpas299 xenograft model with a single intravenous dose of 0.3 or 0.15 mg/kg non-PEGylated ADC:cAC10-[mc-PAB(giuc)-MMAE] _p ( cAC10-1), or its corresponding parallel-oriented PEGylated ADC: cAC10-{mc-[PAB(gluc)-MMAE]-PEG ₂₄ } _p (cAC10-10), where p=4, compared with untreated animals.

Fig. 17. Dose-response curves for hBU12 ADCs loaded with 8 drug groups containing PEGylated scaffolds of different lengths for their PEG groups, against a panel of non-Hodgkin's lymphoma cell lines, where the drug-linker group is represented by the structure MDpr-L ^p (PEG)x( PAB(d1i)), where L ^p represents lysine as a parallel connecting unit, where x = 0 (hBU12-14), where the PEG Group on epsilon-aminolysine is replaced by acetyl, x has a value of 2 (hBU12-43), 4 ( hBU12-42), 8 (hBU12-18), 12 (hBU12-17), 24 (hBU12-16), or is a branched structure of PEG4-(PEG ₄ ) ₃ (hBU12-19).

Fig. 18. Pharmacokinetic profile (total Ab concentration in μg/ml versus time in days) of a rat after a single intravenous 1 mg/kg dose of unconjugated non-targeting antibody (h00), its conjugates containing PEGylated scaffolds of different lengths for its PEG Group, with group drug-linker, represented by the structure MDpr-L ^p -(PEG) _x (PAB(giu)), where L ^p represents lysine as a parallel connecting unit, where x = 0 (h00-14), where PEG Epsilon group -amino lysine is replaced with acetyl, x has a value of 2 (h00-43), 4 (h00-42), 8 (h00-18), 12 (h00-17) or 24 (h00-16).

- 3 046498

Fig. 19. Mean tumor volume (mm ² ) versus days after tumor transplantation in a model of CD19-positive RL diffuse large B-cell lymphoma after intravenous administration of a single dose of 1 or 3 mg/kg hBU12 ADCs containing PEGylated scaffolds of different lengths for their PEG Groups, with drug linker agent represented by the structure MDpr-Lp(Π3Γ) _χ (ΡΑΒ^1^)), where L ^p represents lysine as a parallel connecting unit, where x = 0 (hBU12-14), where PEG Group at epsilon-aminolysine replaced with acetyl, x has a value of 2 (hBU12-43), 4 (hBU12-42), 8 (hBU12-18), 12 (hBU12-17) or 24 (hBU12-16), compared to untreated animals.

Fig. 20. Drug concentrations (nM) in CD30+ L540cy Hodgkin lymphoma tumor xenografts in mice after administration of a single dose of 1 mg/kg non-PEGylated ADC, cACW^mc-PAB^lu^MMAE^, (cAC10-1), parallel-oriented PEGylated ADCs with drug linker mc-I. ^p -(PAB (gluc)-MMAE) PEG24 (cAC10-10), MDpr-L ^p - (PAB (gluc) MMAE)PEG24 (cAC10-16), where the parallel Connecting Unit L ^p is lysine, or series-oriented pegylated ADC (cAC10-4), where the ADCs have an average drug loading of 8.

Fig. 21. Tolerability, shown as % weight change over time, of a single intravenous dose of 50 mg/kg non-targeted control PEGylated drug conjugates containing PEGylated scaffolds of varying lengths for their PEG Group, with a drug linker represented by the MDpr structure -L ^p -(PEG) _x (PAB(gluc)), where I ^p represents lysine as a parallel connecting unit, where x = 0 (h00-43), in which the PEG Group at epsilon-aminolysine is replaced by acetyl, x has value 2 (h00-43), 4 (h0042), 8 (h00-18), 12 (h00-17) or 24 (h00-16), where ADCs have an average drug load of 8, compared to untreated animals.

Fig. 22. Size exclusion chromatography (SEC) chromatograms for non-targeted control PEGylated drug conjugates containing PEGylated scaffolds of varying lengths for their PEG Groups, with a drug linker represented by the MDpr-L ^p -(PEG) _x structure (PAB(gluc)), where L ^p represents lysine as a parallel connecting unit, where x is 8 (h00-18), 12 (h00-17) or 24 (h00-16)

Fig. 23. Half-life and distribution consistent with a two-compartment model for non-targeted control PEGylated drug conjugates containing PEGylated scaffolds of varying lengths for their PEG groups, with a drug linker represented by the MDpr-L ^p -(PEG) _x structure (PAB(gluc)), where L ^p represents lysine as a parallel connecting unit, where x is 8 (h00-18), 12 (h00-17) or 24 (h00-16)

The essence of the invention

The invention provides, among other things, ligand-drug conjugates (LDCs), methods for their preparation and use, and intermediates thereof. Ligand-drug conjugates are stable in the circulation but are capable of causing cell death of target cells or inhibiting the proliferation of target cells once their drug cargo is released near or in the target cells.

In general embodiments, the LDC of the present invention is represented by the structure of Formula I below

drug linker (I) where D represents a drug group, PEG represents a polyethylene glycol group that masks the hydrophobicity of the drug linker, I ^p represents a parallel connecting unit that allows the PEG group to be in a parallel orientation relative to XD, A represents is a branching group when m is greater than 1, optionally consisting of subunits, or A is absent when m is 1, X is a releasable complex group that causes release of each D group from the LDC, and Z is an optional spacer group , through which L ^p is linked to I, which is a targeting ligand.

In other basic embodiments of the LDC of the present invention, a structure is provided

- 4 046498 round formula II below

medicinal, .

linker agent ( ¹ -U where AD represents a drug attachment group that allows for additional attachment of XD components, as indicated by the subscript t, in a parallel orientation relative to the PEG group, and L, Lp, Z, A, X, D, m , p and s have the meanings defined for formula I

In other general embodiments, the LDC of the present invention is represented by the structure of Formula III below

linker drug (111) wherein AD, L, Lp, PEG, Z, A, X, D, m, p, s and t are as defined in formula II.

Description of the invention

general information

The present invention is based, in part, on the surprising discovery that the orientation of the polyethylene glycol moiety (PEG Group) of a ligand-drug conjugate can have a profound effect on the resulting pharmacokinetics of the conjugate. In particular, the present inventors have discovered that parallel placement of the PEG group with respect to the drug group of a ligand-drug conjugate can improve the pharmacokinetics of the conjugate compared to conjugates that either do not contain a PEG group or in which the PEG group is in a serial orientation with drug group. The present inventors have also discovered that the number of polyethylene glycol repeat subunits present in the PEG group influences the resulting pharmacokinetics of the conjugate. By designing conjugates to contain a PEG group in a parallel arrangement and of a suitable size to mask the hydrophobicity of the drug and, in some cases, linker components, it is possible to obtain ligand-drug conjugate formats that allow for higher drug loading while maintaining These are other characteristics of lower loading conjugates, such as favorable pharmacokinetic properties. Ligand-drug conjugates are also designed to release free drug.

Definitions

Unless otherwise specified, the following terms and phrases as used in this application have the following meanings. When trade names are used in this application, the trade name includes the dosage form of the product, the generic drug, and the active pharmaceutical ingredient(s) of the product bearing such trade name, unless the context otherwise requires.

A parallel linker, in the context of the present invention, refers to a branched linker group component that connects the PEG group in a parallel orientation to the drug group. In the context of the present invention, the phrase parallel orientation, parallel arrangement, parallel connection and the like refers to a configuration wherein parallel-arranged or parallel-oriented or parallel-connected components are connected to a parallel connecting link (L ^p ) such that each has one end , connected to Lp, and one free end. Typically Lp connects the drug group through one or more linker group components, of which one (or only) is

- 5046498 is a releasable complex group and a PEG group such that the drug group and the PEG group are in parallel orientation so that the hydrophobicity of the drug group is masked by the PEG group. In some aspects, additional branching is provided by one or more drug attachment groups (ADs) that are linked to the Lp such that the drug group bound to the AD is in a parallel orientation to the PEG group in that Lp. Only those PEG groups that are necessary to mask the hydrophobicity of a given linker-drug group need to be in parallel orientation with their drug group, and there is no need for all of the drug and polyethylene glycol groups bound to L ^p to be in parallel orientation with each other.

The term parallel is used herein to indicate the branching of two components of a ligand-drug conjugate (LDC) from the Lp of the LDC, and is not used to indicate that the two components are located side-by-side in space or have the same distance between them. along some or all of their length. In cases where a parallel-oriented component is itself branched and thus has multiple ends, it nevertheless has only one connected end.

An LDC containing a PEG group that is in a parallel orientation with respect to the drug group of the LDC refers to an LDC comprising a PEG group that has one end that is bonded to a linker group component (i.e., a Parallel Connector), and one or more loose, untied ends. The free unbound end of the PEG group may be in the form of, for example, an unreacted functional group, such as an alkoxy, carboxylic acid, alkylene carboxylic acid, alcohol, or other functional group. The parallel orientation of the PEG Group with respect to the Drug Group minimizes the number of atoms between the Ligand Group and the Drug Group, since the PEG Group atoms are not wedged between the drug Group and the ligand Group. In LDCs, the linker moiety contains a releasable complex moiety capable of releasing the bioactive drug moiety from the LDC at the target site (eg, via intracellular cleavage). In some cases, the drug group that is released is the original drug that was included in the drug group, and thus does not remain associated with the PEG group or degradation product of the ligand group. In other cases, the biologically active drug group that is released is the parent drug containing the remainder of the linker group (other than the PEG group).

A linker group component having a release mechanism, which is indicated as a release complex group, is embedded between L ^p and the drug group. Just as with a PEG group, a drug group has one end that is attached (even if indirectly via a releasable complex group) to a parallel connecting link, and one or more free unconnected ends (or, in the case of some cyclic drugs, no free ends). ends).

An exemplary graphical representation of an LDC containing a PEG group that is in a parallel (i.e., branched) orientation with respect to the drug group is as follows:

C(O)CH ₂ CH2(OCH2CH ₂ )pOCНz „ I

Ligand------- Linker-----Drug

The phrase serial orientation or serial placement or serial connection refers to the configuration of a component in an LDC where a serially oriented component is attached such that it has two associated ends, with those ends each associated with different components of the LDC. An LDC containing a PEG group that is in sequential orientation with respect to the ligand group and the drug group of that LDC refers to an LDC comprising a PEG group that is linked to the ligand at one end (typically indirectly through linker group components) and to the drug group means at the other end (typically indirectly through other components of the linker group). The sequential placement of the PEG group increases the number of atoms between the ligand group and the drug group, since at least some of the atoms of the PEG group are located between the drug group and the ligand group. For example, one or more (OCH2CH2) subunits that characterize the PEG group are inserted between the drug group and the ligand group. An exemplary graphical representation of a Ligand-Drug conjugate containing a PEG group that is in sequential orientation with respect to the ligand group and the drug group is as follows:

Ligand^^OCH^CH^^-drug, where Z1 and Z2 are optional

- 6 046498 extending components of the linker group.

The term antibody, in the context of the present invention, is used in the broadest sense and specifically covers intact monoclonal antibodies, polyclonal antibodies, monospecific antibodies, multispecific antibodies (for example, bispecific antibodies) and antibody fragments that exhibit the desired biological activity, provided that the antibody fragment contains the required number of attachment sites for the drug linker. The native form of the antibody is a tetramer and consists of two identical pairs of immunoglobulin chains, with each pair containing one light chain and one heavy chain. In each pair, the light and heavy chain variable regions (VL and VH) together are primarily responsible for binding to antigen. The light and heavy chain variable regions consist of a framework region interrupted by three hypervariable regions, also called complementarity determining regions or CDRs. Constant regions can be recognized and interact with the immune system. (See, for example, Janeway et al., 2001, Immuno. Biology, 5th Ed., Garland Publishing, New York). The antibody can be of any type (eg, IgG, IgE, IgM, IgD and IgA), class (eg, IgG1, IgG2, IgG3, IgG4, IgA1 and IgA2) or subclass. The antibody may be from any suitable species. In some aspects, the antibody is of human or murine origin. The antibody may be, for example, human, humanized or chimeric.

The term monoclonal antibody, in the context of the present invention, refers to an antibody obtained from a population of essentially homogeneous antibodies, i.e. the individual antibodies that make up the population are identical, except for possible natural mutations that may be present in minor quantities. Monoclonal antibodies are highly specific, directed against a single antigenic site. The designation monoclonal indicates the characteristic of an antibody as being derived from an essentially homogeneous population of antibodies, and should not be construed as requiring the antibody to be produced by any particular method.

An intact antibody is one that includes an antigen-binding variable region as well as a light chain constant region ( _CL ) and heavy chain constant regions, CH1, CH2, CH3 and CH4, depending on the class of the antibody. The constant regions may be native sequence constant regions (eg, human native sequence constant regions) or an amino acid sequence variant thereof.

An antibody fragment includes a portion of an intact antibody including its antigen binding or variable region. To be useful in the present invention, the antibody fragment must contain the required number of sites for attachment to the linker drug. Attachment sites may be of natural or non-natural origin.

An antigen is a unit to which an antibody binds in a specific manner.

The terms specific binding and specifically binds mean that the antibody or antibody derivative will bind, in a highly selective manner, to its respective target antigen and not to a large number of other antigens. Typically, the antibody or antibody derivative binds with an affinity of at least about 1x 10 ^-7 M, and preferably from 10 ^-8 M to 10 ^-9 M, 10 ^-10 M, 10 ^-11 M or 10 ^-12 M, and binds to pre- a specific antigen with an affinity that is at least two times greater than its affinity for binding to a nonspecific antigen (eg, BSA, casein) other than a predetermined antigen or a closely related antigen.

The term inhibit or inhibit means to reduce by a measurable amount or completely prevent.

The term therapeutically effective amount refers to an amount of a conjugate effective to treat a disease or disorder in a mammal. In the case of cancer, a therapeutically effective amount of the conjugate can reduce the number of cancer cells; reduce tumor size; inhibit (ie, slow to some extent and preferably stop) the infiltration of cancer cells into peripheral organs; inhibit (ie, slow to some extent and preferably stop) tumor metastasis;

inhibit, to some extent, tumor growth; and/or alleviate to some extent one or more symptoms associated with cancer. To the extent that a drug can inhibit the growth and/or kill existing cancer cells, it may be cytostatic and/or cytotoxic. For cancer therapy, efficacy, for example, can be measured by determining the time to disease progression (TTP) and/or determining the percentage of positive clinical response (RR).

Unless the context otherwise requires, the term essential or essentially refers to a major part, i.e. >50% of a population, mixture or sample, preferably greater than 50%, 55%, 60%, 65%, 70%, 75%, 80%, 85%, 90%, 91%, 92%, 93%, 94% , 95%, 96%, 97%, 98% or 99% of the population.

The terms intracellularly degradable and intracellular degradation refer to metabo

- 7 046498 personal process or reaction within a cell to a ligand-drug conjugate (for example, an antibody-drug conjugate (ADC) or the like), resulting in a covalent attachment, between the drug group (D) and the ligand group ( e.g., antibody (Ab)) is degraded, for example, by the action of the released complex group, leading to the dissociation of free drug from the LDC, including its degradation products, within the cell. The substances resulting from such dissociation are thus intracellular metabolites.

The term cytotoxic activity refers to the cell killing effect of a drug or ligand-drug conjugate or an intracellular metabolite of a ligand-drug conjugate.

Cytotoxic activity can be expressed by the IC50 value, which is the concentration (molar or mass) per unit volume at which half of the cells survive exposure to a cytotoxic agent.

The term cytostatic activity refers to an antiproliferative effect other than cell killing, a cytostatic agent, or a ligand-drug conjugate containing a cytostatic agent as its drug moiety, or an intracellular metabolite thereof, where the metabolite is a cytostatic agent.

The term cytotoxic agent, in the context of the present invention, refers to a substance that has cytotoxic activity and causes cell destruction. The term is intended to include radioactive isotopes (e.g. ²¹¹ At, ¹³¹ I, ¹²⁵ I, ⁹⁰ Y, ¹⁸⁶ Re, ¹⁸⁸ Re, ¹⁵³ Sm, ²¹² Bi, ³² P, ⁶⁰ C and Lu radioactive isotopes), chemotherapeutic agents and toxins such as low molecular weight toxins or enzymatically active toxins of bacterial, fungal, plant or animal origin, including their synthetic analogues and derivatives.

The term cytostatic agent, in the context of the present invention, refers to a substance that has cytostatic activity, for example, inhibiting the function of cells responsible for or promoting cell growth or reproduction. Cytostatic agents include inhibitors, such as protein inhibitors, eg enzyme inhibitors.

The terms cancer and cancerous refer to a physiological condition or disorder in mammals or describe a condition or disorder that is typically characterized by unregulated cell growth. A tumor contains one or more cancer cells.

An autoimmune disease, as used herein, is a disease or disorder arising from, and directed against, an individual's own tissues or proteins.

A patient, in the context of the present invention, refers to a subject to whom LDC is administered. Examples of a patient include, but are not limited to, a human, rat, mouse, guinea pig, non-human primate, pig, goat, cow, horse, dog, cat, bird, or game. Typically, the patient is a rat, mouse, dog, non-human primate or human. In some aspects, the patient is a person in need of an effective amount of LDC.

The terms treat or cure, unless the context otherwise requires, refer to therapeutic treatment and prophylactic measures to prevent relapse, where the goal is to inhibit or slow (reduce) an undesirable physiological change or disorder, such as, for example, the development or spread of cancer. For purposes of the present invention, beneficial or desirable clinical results include, but are not limited to, relief of symptoms, reduction in the severity of disease, stabilization (i.e., not worsening) of a disease state, delay or slowing of disease progression, reduction in severity or temporary relief of disease, and remission (partial or complete), regardless of whether it is detectable or not. Treatment may also mean longer survival than would be expected without treatment. Subjects in need of treatment include those who already have such a condition or disorder, as well as those who are predisposed to such a condition or disorder.

In the context of cancer, the term treatment includes any or all of the following: inhibition of the growth of tumor cells, cancer cells or tumors; inhibiting the replication of tumor cells or cancer cells, reducing the overall tumor burden or reducing the number of cancer cells and reducing the severity of one or more symptoms associated with the disease.

In the context of an autoimmune disease, the term treatment includes any or all of the following: inhibiting the replication of cells associated with the autoimmune disease, including, but not limited to, cells that produce the autoimmune antibody, reducing the load of autoimmune antibodies, and reducing the severity of one or more symptoms of the autoimmune disease .

The phrase pharmaceutically acceptable salt, as used herein, refers to pharmaceutically acceptable organic or inorganic salts of a compound (eg, drug, drug-linker, or ligand-drug conjugate). The compound may contain at least one amino group and, accordingly, acidic

- 8 046498 addition salts can be formed with the amino group. Examples of salts include, but are not limited to, sulfate, trifluoroacetate, citrate, acetate, oxalate, chloride, bromide, iodide, nitrate, bisulfate, phosphate, acid phosphate, isonicotinate, lactate, salicylate, acid citrate, tartrate, oleate, tannate, pantothenate , bitartrate, ascorbate, succinate, maleate, gentisinate, fumarate, gluconate, glucuronate, sucrose, formate, benzoate, glutamate, methanesulfonate, ethanesulfonate, benzenesulfonate, p-toluenesulfonate and pamoate (i.e. 1,1'-methylene-bis- (2-hydroxy-3-naphthoate)). The pharmaceutically acceptable salt may include the inclusion of another molecule, such as an acetate ion, succinate ion, or other counterion. The counterion can be any organic or inorganic group that stabilizes the charge on the parent compound. In addition, a pharmaceutically acceptable salt may contain more than one charged atom in its structure. Cases where multiple charged atoms are part of a pharmaceutically acceptable salt may involve multiple counterions. Therefore, a pharmaceutically acceptable salt may contain one or more charged atoms and/or one or more counterions.

Unless otherwise specified, the term alkyl, as such or as part of another term, refers to a substituted or unsubstituted, straight or branched, saturated or unsaturated hydrocarbon containing the indicated number of carbon atoms (for example, -C ₁ -C ₈ alkyl or -C ₁ C ₁₀ alkyl refers to an alkyl group containing 1 to 8 or 1 to 10 carbon atoms, respectively). When the number of carbon atoms is not specified, the alkyl group contains from 1 to 8 carbon atoms. Representative linear -C1-C8 alkyl groups include, but are not limited to, -methyl, -ethyl, -n-propyl, -n-butyl, -n-pentyl, -n-hexyl, -n-heptyl and -n-octyl ; while branched -C1-C8 alkyls include, but are not limited to, -isopropyl, -sec-butyl, -isobutyl, -tert-butyl, -isopentyl and -2-methylbutyl; unsaturated -C ₂ -C ₈ alkyls include, but are not limited to, -vinyl, -allyl, -1-butenyl, -2-butenyl, isobutylenyl, -1-pentenyl, -2-pentenyl, -3-methyl-1- butenyl, -2methyl-2-butenyl, -2,3-dimethyl-2-butenyl, -1-hexyl, 2-hexyl, 3-hexyl, -acetylenyl, -propynyl, -1-butynyl, -2-butynyl, -1- pentynyl, -2-pentynyl and -3-methyl-1-butynyl. In some aspects, the alkyl group is unsubstituted. The alkyl group may be substituted by one or more groups. In other aspects, the alkyl group will be saturated.

Unless otherwise specified, alkylene, as such or as part of another term, refers to a substituted or unsubstituted, saturated or unsaturated, branched or linear or cyclic hydrocarbon radical containing a specified number of carbon atoms, typically 110 carbon atoms, and containing two monovalent radical centers , formed by removing two hydrogen atoms from the same or two different carbon atoms of the parent alkane. Typical alkylene radicals include, but are not limited to: methylene (-CH ₂ -), 1,2-ethyl (-CH ₂ CH ₂ -), 1,3-propyl (-CH ₂ CH ₂ CH ₂ -), 1, 4-butyl (-CH ₂ CH ₂ CH ₂ CH ₂ -), etc. In preferred aspects, the alkylene is a branched or linear hydrocarbon (ie, it is not a cyclic hydrocarbon). In any of the embodiments presented herein, the alkylene may be a saturated alkylene.

Unless otherwise specified, aryl, as such or as part of another term, means a substituted or unsubstituted monovalent carbocyclic aromatic hydrocarbon radical containing 6-20 carbon atoms (preferably 6-14 carbon atoms), formed by the removal of one hydrogen atom from one carbon atom the original aromatic ring system. Some aryl groups are represented in the exemplary structures as Ar. Typical aryl groups include, but are not limited to, radicals derived from benzene, substituted benzene, naphthalene, anthracene, biphenyl, and the like. An example of an aryl group is a phenyl group.

Unless otherwise specified, arylene, as such or as part of another term, is an aryl group as defined above, where one of the hydrogen atoms of the aryl group is replaced by a bond (i.e., divalent), and may be ortho, meta, or para orientations as shown in the following structures, and an illustrative group is phenyl:

In selected embodiments, for example, when the parallel linking unit, branching group, or drug attachment group includes an arylene, the arylene group is an aryl group as defined above wherein one or two of the hydrogen atoms of the aryl group are replaced by a bond (i.e., the arylene may be divalent or trivalent).

Unless otherwise specified, a C ₃ -C ₈ heterocycle, as such or as part of another term, refers to a monovalent substituted or unsubstituted, aromatic or non-aromatic, monocyclic or bicyclic ring system containing from 3 to 8 carbon atoms (also called ring rings). members) and from one to four heteroatoms as ring members, regardless of

- 9 046498 branes of N, O, P or S, and formed by removing one hydrogen atom from the ring atom of the original ring system. One or more N, C or S atoms in the heterocycle may be oxidized. The ring that includes the heteroatom may be aromatic or non-aromatic. Unless otherwise noted, a heterocycle is linked to its side group at any heteroatom or carbon atom, resulting in a stable structure. Representative examples of the C ₃ -C ₈ heterocycle include, but are not limited to, pyrrolidinyl, azetidinyl, piperidinyl, morpholinyl, tetrahydrofuranyl, tetrahydropyranyl, benzofuranyl, benzothiophene, indolyl, benzopyrazolyl, pyrrolyl, thiophenyl (thiophene), furanyl, thiazolyl, imidazolyl, pyrazolyl, pyrimidinyl, pyridinyl, pyrazinyl, pyridazinyl, isothiazolyl and isoxazolyl.

Unless otherwise specified, C3-C8 heterocyclo, as such or as part of another term, refers to a _C3 - _C8 heterocyclic group as defined above, where one of the hydrogen atoms of the heterocyclic group is replaced by a bond (ie, it is divalent). In selected embodiments, for example, when the parallel linking unit, branching group, or drug attachment group includes a heterocyclo, the heterocyclo is a heterocyclic group as defined above, wherein one or two of the hydrogen atoms of the heterocyclic group are replaced by a bond (i.e., the heterocyclo may be divalent or trivalent).

Unless otherwise specified, a C ₃ -C ₈ carbocycle, as such or as part of another term, is a 3-, 4-, 5-, 6-, 7-, or 8-membered monovalent substituted or unsubstituted, saturated or unsaturated un- an aromatic monocyclic or bicyclic carbocyclic ring formed by removing one hydrogen atom from the ring atom of the original ring system. Representative -C ₃ -C ₈ carbocycles include, but are not limited to, cyclopropyl, cyclobutyl, cyclopentyl, cyclopentadienyl, cyclohexyl, cyclohexenyl, 1,3-cyclohexadienyl, 1,4-cyclohexadienyl, cycloheptyl, 1,3-cycloheptadienyl, 1,3 ,5-cycloheptatrienyl, cyclooctyl and cyclooctadienyl.

Unless otherwise specified, C ₃ -C ₈ carbocyclo, as such or as part of another term, refers to a C ₃ -C ₈ carbocyclic group as defined above, where one of the hydrogen atoms of the carbocyclic group is replaced by a bond (i.e., it is divalent ). In selected embodiments, for example, when the parallel linking unit, branching group, or drug attachment group includes a carbocyclo, the carbocyclo is a carbocyclic group as defined above, wherein one or two of the hydrogen atoms of the carbocyclic group are replaced by a bond (i.e., the carbocyclo may be divalent or trivalent).

Unless otherwise specified, the term heteroalkyl, alone or in combination with another term, means, unless otherwise specified, a stable hydrocarbon, straight or branched chain or combination thereof, fully saturated or containing from 1 to 3 degrees of unsaturation, consisting of a specified amount carbon atoms and from one to ten, preferably from one to three heteroatoms selected from the group consisting of O, N, Si and S, and wherein the nitrogen and sulfur atoms may optionally be oxidized, and the nitrogen heteroatom may optionally be quaternized. The heteroatom(s) O, N and S can be at any internal position of a heteroalkyl group or at a position at which an alkyl group is attached to the rest of the molecule. The Si heteroatom can be located at any position of the heteroalkyl group, including the position at which the alkyl group is attached to the rest of the molecule. Examples include -CH2-CH2-O-CH3, -CH2-CH2-NH-CH3, -CH2-CH2-N( _CH3 ) _-CH3 , -CH2-S-CH2-CH3, _-CH2 - _CH2- S(O)-CH ₃ , -NH-CH2-CH2-NH-C (O)-CH ₂ -CH ₃ , -CH ₂ -CH ₂ -S(O) ₂ -CH ₃ , -CH=CH-O -CH ₃ , -Si(CH ₃ ) ₃ , -CH ₂ -CH=NO-CH ₃ and -CH=CH-N(CH ₃ ) -CH3. Up to two heteroatoms can be arranged in series one after another, for example -CH ₂ -NH-OCH3 and -CH ₂ -O-Si(CH ₃ ) ₃ . In preferred embodiments, the C1-C4 heteroalkyl or heteroalkylene contains from 1 to 4 carbon atoms and 1 or 2 heteroatoms, and the Ci- _C3 heteroalkyl or heteroalkylene contains from 1 to 3 carbon atoms and 1 or 2 heteroatoms. In some aspects, heteroalkyl or heteroalkylene is saturated.

Unless otherwise specified, the term heteroalkylene, alone or as part of another substituent, means a divalent group derived from heteroalkyl (discussed above), and examples include -CH2-CH2-S-CH2-CH2- and -CH2-S- CH2-CH2-NH-CH2-. As for heteroalkylene groups, heteroatoms are also present at either or both ends of the chain. In addition, for alkylene and heteroalkylene linking groups, no orientation of the linking group is implied. In selected embodiments, for example, when the parallel linking unit, branching group, or drug attachment group includes a heteroalkylene, the heteroalkylene is a heteroalkylene group as defined above, wherein one or two of the hydrogen atoms of the heteroalkyl group are replaced by a bond (i.e., the heteroalkylene may be divalent or trivalent).

Substituted alkyl and substituted aryl mean alkyl and aryl, respectively, in which one or more hydrogen atoms are each independently substituted by a substituent. Typical substituents include, but are not limited to, -X, -R, ^-O- , -OR, -SR, ^-S- , -NR2, -NR3, =NR, -CX3, -CN, -OCN, -SCN, -N=C=O, -NCS, -NO, -NO2, =N2, -N3, -NRC(=O)R, -C(=O)R, -C(=O)NR2, -SO3 ^- , -SO3H, -S(=O)2R, -OS(=O) ₂ OR, -S(=O) ₂ NR, -S(=O)R, -OP (=O)(OR) ₂ , P( =O)(OR) ₂ , -PO ^- ₃ , -PO ₃ H ₂ , -AsO ₂ H ₂ , -C(=O)R, -C(=O)X,

- 10 046498

-C(=S)R, -CO2R, -CO2 ^- , -C(=S)OR, C(=O)SR, C(=S)SR, C(=O)NR2, C(=S)NR2 or C(=NR)NR2, where each X independently represents a halogen: -F, -Cl, -Br or -I; and each R independently represents -H, -C ₁ -C _2O alkyl, -C ₆ -C ₂₀ aryl, -C ₃ -C ₁₄ heterocycle, protecting group or prodrug group. Typical substituents also include (=O). The alkylene, carbocycle, carbocyclo, arylene, heteroalkyl, heteroalkylene, heterocycle and heterocyclo groups described above may also be similarly substituted.

As used herein, the term free drug refers to a biologically active drug moiety that is not covalently attached, either directly or indirectly, to a PEG moiety or a degradation product of a ligand moiety. Free drug may refer to the drug as it exists immediately after cleavage from the linker group through the release mechanism provided by the release complex group in the LDC, or to subsequent intracellular conversion or metabolism. In some aspects, the free drug will be in the H-D form or may exist as a charged group. A free drug is a pharmacologically active compound that can provide the desired biological effect. In some aspects, the pharmacologically active compound is not the parent drug, but may include a linker moiety component that has not undergone subsequent intracellular metabolism.

Ligand-drug conjugate compounds and corresponding intermediates

The present invention is based, in part, on the discovery that for ligand-drug conjugates (LDCs) that have unfavorable pharmacokinetic properties, these pharmacokinetic properties can be improved by placing the PEG groups in a parallel orientation relative to the drug group, as described herein. In some aspects, the clearance profile of PEGylated conjugates is the same as that of an unconjugated ligand (ie, a targeting agent such as an antibody or corresponding antigen-binding moiety), even at high drug levels. LDCs include a ligand group (ie, a targeting ligand), a linker group, and a drug group. The linker group, before or after its attachment to the targeting ligand, connects the drug group to the ligand group and includes a PEG group in a parallel configuration to the drug group. This parallel configuration results from the attachment of the drug group, through the release complex group, and the PEG group to the parallel connecting link. A linker group, when linked to a drug group, may be referred to as a drug linker. The population of LDCs will preferably have an average loading of drug-linker groups of at least about 6, about 7, or about 8 drug-linker groups per ligand group.

The PEG groups are designed to provide an optimal level of masking of the hydrophobicity of the hydrophobic drug linker components. Therefore, the inclusion of a PEG moiety as disclosed herein is particularly suitable for linker drugs that would otherwise be sufficiently hydrophobic to negatively impact the pharmacokinetics of the resulting conjugate compared to the unconjugated ligand. These lower pharmacokinetic properties include higher plasma clearance. Thus, ligand-drug conjugates that exhibit significantly higher plasma clearance and correspondingly lower plasma drug content compared to unconjugated Ligand will be useful in the present invention.

Ligand-drug conjugates have more favorable pharmacokinetic properties due to the parallel orientation of the drug moiety and the PEG moiety in the hydrophobic drug-linker moiety, whereby the negative influence of the hydrophobicity of the drug moiety and/or other components of the drug-linker moiety on plasma clearance is reduced or is eliminated (i.e., the hydrophobicity of the drug-linker portion is masked). Parallel orientation is provided by the parallel link link (Lp), since the parallel link link acts to connect the drug group, the PEG group and the ligand in a suitable branching configuration, providing the required parallel orientation. The parallel linker can be considered a scaffold containing attachment sites for the conjugate components, which can be multiplexed to have multiple drug groups in parallel orientation with the PEG groups to provide a PEGylated multiplex scaffold. In some embodiments, the hydrophobic moiety in the drug linker portion, the hydrophobicity of which is masked by the parallel-oriented PEG group, is a hydrophobic drug moiety.

The drug group is attached to the parallel connecting link through a releasable complex group. The released complex group allows for effective release of the drug from the target cell sufficient to cause, for example, cytotoxicity or cytostaticity. Typically, the release complex moiety is designed to effectively release free drug once the conjugate is internalized into the target cell, but is also designed to release free drug in close proximity to the target cells. Suitable recognition sites for cleavage are those that allow efficient release of the drug group(s) contained in the LDC. Typically, the recognition site is a peptide cleavage site (such as in peptide-based release complex groups), a sugar cleavage site (such as in sugar-based release complex groups), or a disulfide cleavage site (such as in disulfide-based release complex groups) . Examples of peptide cleavage sites include those recognized by intracellular proteases, such as those found in lysosomes. Examples of sugar cleavage sites include sites recognized by glycosidases, including glucuronidases such as beta-glucuronidase.

Any bioactive compound (ie, drug) can be used as a drug group in the present invention. The bioactive compound may contain a suitable site for inclusion as a drug group in an LDC or may be modified for this purpose while substantially retaining the desired biological activity of the parent drug when the modified drug, which may or may not retain part of the linker group , is released from LDC. Preferred drug groups provide release of the parent bioactive compound. The drug group may be an auristatin or non-austatin drug, which is a hydrophobic component of the drug-linker portion, the hydrophobicity of which should be masked by a parallel-oriented drug group. The effects of the present invention will be more pronounced in embodiments where the drug moiety, release complex moiety, or drug moiety/release complex moiety combination are hydrophobic in nature, thereby negatively affecting the pharmacokinetics of the resulting conjugate. Examples of hydrophobic drugs include monomethyl auristatin E and drugs having a hydrophobicity that is comparable to or greater than that of monomethyl auristatin E. Examples of hydrophobic Release Complexes include peptide- and sugar-based Release Complexes that contain a hydrophobic self-cleaving component specifically illustrated herein application, as well as the Released Complex Groups, having a hydrophobicity comparable to or greater than that of such Released Complex Groups.

Hydrophobicity can be measured using SlogP. SlogP is defined as the log of the octanol/water partition coefficient (including implied hydrogens), and can be calculated using the MOE™ program from the Chemical Computing group (SlogP values calculated according to Wildman, S.A., Crippen, G.M.; Prediction of Physiochemical Parameters by Atomic Contributions; J. Chem. Inf. Comput. Sci. 39 No. 5 (1999) 868-873). When reference is made to a drug group or release complex group having a hydrophobicity comparable to the reference drug group or release complex group, the SlogP value will be within 20%, preferably within 10%, of the SlogP value of the reference drug group or release complex group .

In light of the foregoing, the present invention provides, in one set of embodiments, a composition of ligand-drug conjugates comprising a population of ligand-drug conjugates. Ligand-drug conjugates include a ligand group and several drug-linker groups attached to it. Preferably, on average there are from about 6 to about 14, from about 6 to about 12, from about 6 to about 10, from about 8 to about 14, from about 8 to about 12, from about 8 to about 10 drug-linker groups on the Ligand in the composition. An example of attachment to a ligand is attachment via thioether bonds. Examples of conjugation sites on a ligand are thiol groups resulting from the reduction of interchain disulfide residues and/or thiol-containing residues embedded in the ligand, such as embedded cysteines. The attachment can be, for example, through thiol residues formed from the interchain disulfide and from 0-8 built-in cysteine residues.

In a related set of embodiments, methods are provided for administering ligand-drug conjugates to a patient to treat a disease. The disease may be, for example, cancer or an autoimmune disease.

Ligand-drug conjugates are administered in a therapeutically effective amount and in a therapeutically effective schedule.

Embodiments

Various embodiments of the invention are described below, followed by a more detailed discussion of the components for preparing ligand-drug conjugates and their intermediates.

- 12 046498 connections. Any of the selected embodiments for the components of ligand-drug conjugates and intermediates thereof may relate to any and all aspects of the invention described herein, or they may relate to one aspect. Individual embodiments may be combined together in any combination.

Ligand-drug conjugate compounds

In one set of embodiments, LDC compounds capable of releasing free drug are provided, wherein the LDC compound is represented by Formula AA below:

a linker drug (ΆΑ) or a pharmaceutically acceptable salt of such a compound, wherein

L represents a ligand group;

D represents a drug group;

PEG is a polyethylene glycol group;

Z represents an expansion group;

X represents a releasable complex Group;

Lp represents a parallel connecting link;

A represents an optional branching group;

the subscript p is an integer having a value from 1 to 14, preferably from 2 to 12 (preferably from 6 to 14, from 6 to 12, from 8 to 14, or from 8 to about 12);

subscript m is an integer ranging from 1 to 4; and preferably has a value of 1 or 2; and the subscript s has the value 0 or 1, with the proviso that when s has the value 0, m has the value 1, and when s has the value 1, m has the value 2, 3, or 4.

In another group of embodiments, Formula AA does not represent individual LDC compounds, but an LDC composition (ie, a composition including a population of individual LDC compounds). In such embodiments, p represents the average number of drug linkers per ligand in the composition. In such embodiments, p is typically not an integer value and can range from 1 to about 14, preferably from about 2 to about 12 (preferably from about 6 to about 14, from about 6 to about 12, from about 8 to about 14 or from about 8 to about 12). Other variables (eg L, Z, A, Lp, PEG, X, D, s and m) have the meaning defined above.

In another group of embodiments, the LDC composition includes a population of LDC compounds, wherein the individual LDC compounds are represented by the formula AA, wherein for each individual LDC compound, p is independently selected from an integer having a value of from 1 to 14, preferably from 2 to 12 (preferably from 6 to 14, 6 to 12, 8 to 14, or 8 to about 12), and the average number of drug linkers per ligand in the composition is from 1 to about 14, preferably from about 2 to about 12 (preferably from about 6 to about 14, about 6 to about 12, about 8 to about 14, or about 8 to about 12).

In some aspects, from 1 to 32 or from 2 to 32 (preferably from 6 to 32 or from 8 to 32) drug groups are attached to each ligand group. The population of ligand-drug conjugates may contain an average of 1 to 32 or about 2 to 32 (preferably about 6 to 32 or about 8 to 32) drug groups per ligand.

Specific embodiments of LDC compounds or LDC compositions represented by Formula AA include those in which:

1) m has the value 1 and s has the value 0;

2) m has a value from 2 to 4, and s has a value of 1;

3) m has the value 2 and s has the value 1;

4) m has the value 1; s has the value 0; and p is an integer having a value of 6 to 14, 8 to 14, or 8 to 12 for an LDC compound, or p is a number having a value of 6 to about 14, about 8 to about 14, or about 8 to about 12 for LDC composition;

5) m has a value of 2-4; s has the value 1 and p is an integer having a value from 6 to 14, 8 to 14 or 8 to 12 for an LDC connection, or p is a number having a value from 6 to about 14, from about 8 to about 14 or about 8 to about 12 for an LDC composition;

- 13 046498

6) m has the value 2; s has the value 1; and p is an integer having a value of 6 to 14, 8 to 14, or 8 to 12 for an LDC connection; or p is a number having a value of from 6 to about 14, from about 8 to about 14, or from about 8 to about 12 for an LDC composition;

7) m has the value 2; s has the value 1; and p has a value of 8

8) m has the value 1; s has the value 0; and p has a value of 8

9) Any of the embodiments described in paragraphs 1 to 8 of this subsection, wherein from 1 to 32 or about 2 to 32 (preferably from about 6 to about 32 or from about 8 to about 32) drug groups are attached to the ligand group .

10) Any of the embodiments described in paragraphs 1-9 of this subsection, where Lp is a natural or non-naturally occurring amino acid, amino alcohol, aminoaldehyde or polyamine.

Certain embodiments of LDC compounds or LDC compositions that are represented by formula AA have formulas AA1 and AA2, shown below:

or a pharmaceutically acceptable salt of such a compound, wherein

L represents a ligand group;

D represents a drug group;

PEG is a polyethylene glycol group;

Z represents an expansion group;

X represents a releasable complex group;

L ^p represents a parallel connecting link;

A represents a branching group that is present; and the subscript p is an integer having a value of 1 to 14, and preferably 2 to 12 (preferably 6 to 14, 6 to 12, 8 to 14, or 8 to 12) for the ligand-conjugate compound drug, or p is a number ranging from 1 to about 14, and preferably from about 2 to about 12 (preferably from about 6 to about 14, from about 6 to about 12, from about 8 to about 14, or from about 8 to about 12) for a ligand-drug conjugate composition.

In any of the selected embodiments for LDC compounds provided herein where p is present, including the embodiments described above, p may be an integer having a value of 1 to 14, 2 to 14, 2 to 10 , from 4 to 12, from b to 14, from b to 12, from 8 to 12 or from 8 to 10. The index p can have the value 1 or 2, or 3, or 4, or 5, or 6, or 7 or 8, or 9, or 10, or 11, or 12, or 13, or 14.

In any of the selected embodiments for LDC compositions provided herein where a p value is present, including the embodiments described above, p has a value of 1 to about 14, about 2 to about 14, about 2 to about 10, from about 4 to about 12, from about 6 to about 14, from about 6 to about 12, from about 8 to about 12, or from about 8 to about 10. The index p may have a value of 1 or about 1, or 2 or about 2 , or 3 or about 3, or 4 or about 4, or 5 or about 5, or 6 or about 6, or 7 or about 7, or 8 or about 8, or 9 or about 9, or 10 or about 10, or 11 or about 11, or 12 or about 12, or 13 or about 13, or 14 or about 14.

In another set of embodiments, ligand-drug conjugates (LDCs) are provided capable of releasing free drug, wherein from one to thirty-two drug groups (preferably from 2 to 32 drug groups, from 6 to 32 drug groups, from 8 up to 32 drug groups, 6 to 14 drug groups, about 8 to about 14 drug groups, or about 8 to about 12 drug groups) are conjugated to a targeting ligand in the LDC via linker groups, wherein each drug group of the drug in the drug-linker moiety is attached to its linker group through a cleavable moiety (i.e., a releasable complex group) that releases free drug in the vicinity of the ligand-targeted site (L), and wherein the LDCs also include a parallel linking unit (Lp), to which a ligand group is bonded, and a polyethylene glycol (PEG) group, wherein the PEG and drug groups in the linker-drug portion are bonded in a parallel orientation to each other. The Polyethylene glycol group contains from 4 to 72 (preferably from 6 to 72 repeating -OCH ₂ CH ₂ - units, more preferably from 6 to 36 or from 8 to 24) repeating units. The ligand may be an antibody moiety, preferably an intact antibody moiety. The cleavable linker may include, for example, a peptide cleavage site, a ca cleavage site

- 14 046498 hara or disulfide cleavage site. The drug may be an auristatin or a non-austatin. Auristatin or non-auristatin may have a hydrophobicity that is comparable to or greater than monomethyl auristatin E. Auristatin may be monomethyl auristatin E. In some aspects, the ADC exhibits superior pharmacokinetic properties compared to the same or substantially the same ADC not containing a PEG group or containing a PEG group, but in a sequential orientation with respect to the antibody and drug. In some aspects, the ADC exhibits pharmacokinetic properties that are the same or substantially the same as the antibody when it is not a conjugated moiety.

Drug-linker compounds

In some aspects, when constructing ligand-drug conjugates, it is desirable to synthesize the entire linker drug prior to conjugation to the Ligand group. In such embodiments, the drug-linker compounds act as intermediates. Examples of drug-linker compounds are presented below and their structure is represented by the formula BB

or a pharmaceutically acceptable salt of such a compound, wherein D represents a drug group; PEG is a polyethylene glycol group; Z' represents an extending group capable of forming a covalent attachment to a ligand group;

X represents a releasable complex group;

Lp represents a parallel connecting link;

A represents an optional branching group;

subscript m is an integer ranging from 1 to 4; and preferably has a value of 1 or 2;

subscript s has the value 0 or 1, provided that when s has the value 0, m has the value 1, and when s has the value 1, m has the value between 2 and 4.

Certain embodiments of the BB formula include those in which:

1) m has the value 1 and s has the value 0;

2) m has the value 2, 3 or 4 and s has the value 1;

3) m has the value 2 and s has the value 1;

4) Any of the embodiments described in paragraphs 1-3 of this subsection, where Lp is a natural or non-naturally occurring amino acid, amino alcohol, aminoaldehyde or polyamine.

Some embodiments of the BB formulas include the following formulas:

or a pharmaceutically acceptable salt of such a compound, wherein

D represents a drug group;

PEG is a polyethylene glycol group;

Z' represents an extending group capable of forming a covalent attachment to a ligand group;

X represents a releasable complex group;

L ^p represents a parallel connecting link and

A represents a branching group that is present.

Linker intermediates

In some aspects, when constructing ligand-drug conjugates, it may be desirable to conjugate linker components to a ligand group (eg, antibody) prior to attaching the -X-D component of the ligand-drug conjugate. For example, in embodiments where a thiol-containing substituent, eg, cysteine, is used to attach the X-D moiety, it may be desirable to conjugate the linker components to a ligand group (eg, antibody) prior to attaching the -X-D moiety of the ligand-drug conjugate. In some such embodiments, the parallel linking unit is capable of forming a covalent bond with the releasable complex group, but is not yet attached to it. The parallel connecting link may be protected with protecting groups to simplify synthesis. Protection

- 15 046498 This group can be removed immediately before joining the released complex group.

The following are examples of linker intermediates having the formula CC:

or a pharmaceutically acceptable salt of such a compound, wherein

PEG is a polyethylene glycol group;

Z' represents an extending group capable of forming a covalent attachment to a Ligand Group;

A represents an optional branching group;

L ^P represents a parallel connecting unit capable of forming a covalent attachment to a drug-releasing group;

subscript m is an integer ranging from 1 to 4; and preferably has a value of 1 or 2; and the subscript s has the value 0 or 1, with the proviso that when s has the value 0, m has the value 1, and when s has the value 1, m has the value 2, 3, or 4.

Certain embodiments of the CC formula include the following formulas:

or a pharmaceutically acceptable salt of such a compound, wherein PEG represents a polyethylene glycol group;

Z' represents an extending group capable of forming a covalent attachment to a Ligand Group;

-X-D is a releasable complex group attached to a drug group;

A represents a branching group and

L ^P is a parallel connecting unit capable of forming a covalent attachment to -XD.

In some aspects, linker intermediates will be conjugated to a ligand group to form ligand-linker intermediates. Exemplary embodiments of ligand-linker intermediates are represented by the structure shown below:

or a pharmaceutically acceptable salt of such compound where

L represents a ligand group;

PEG is a polyethylene glycol group;

Z represents an expansion group;

L ^P is a parallel connecting unit capable of forming a covalent attachment to -XD;

A represents an optional branching group;

the subscript p is an integer having a value from 1 to 14, preferably from 2 to 12 (preferably from 6 to 14, from 6 to 12, from 8 to 14 or from 8 to 12);

subscript m is an integer ranging from 1 to 4; preferably 1 or 2 and the subscript s has the value 0 or 1, with the proviso that when s has the value 0, m has the value 1, and when s has the value 1, m has the value 2, 3 or 4.

In another set of embodiments, formula DD does not represent individual ligand-linker intermediates, but rather a composition comprising a population of individual ligand-linker intermediates. In such embodiments, p is the mean

- 16 046498 number of intermediate linkers to the ligand in the composition. In such embodiments, p is typically not an integer value and can range from 1 to about 14, preferably from about 2 to about 12 (preferably from about 6 to about 14, from about 6 to about 12, from about 8 to about 14 or from about 8 to about 12). Other variables (eg L, Z, A, LP, PEG, s and m) have the meaning defined above.

Certain embodiments of the DD formula include the following formulas.

or a pharmaceutically acceptable salt of such compound where

L represents a ligand group;

PEG is a polyethylene glycol group;

Z- represents an extension group;

-X-D is a releasable complex group attached to a drug group;

LP is a parallel connecting unit capable of forming a covalent attachment to -X-D;

A represents a branching group and the subscript p is an integer having a value from 1 to 14, preferably from 2 to 12 (preferably from 6 to 14, from 6 to 12, from 8 to 14 or from 8 to 12) for intermediate The ligand linker of a compound, or subscript p, is a number ranging from 1 to about 14, preferably from about 2 to about 12 (preferably from about 6 to about 14, from about 6 to about 12, from about 8 to about 14, or from about 8 to about 12) for the intermediate ligand-linker composition.

Additional embodiments

Conjugates of formula AA and their intermediates provide the ability to include one drug group per PEG group, 1:1 ratio. It may be desirable, however, to provide drug conjugates containing either 1 drug per PEG group or 2 or more drugs per PEG group. Accordingly, the present invention provides ligand-drug conjugates containing at least one drug per PEG group, and intermediates thereof.

It will be appreciated by those skilled in the art that, provided the essential components of the ligand-drug conjugates are present (i.e., ligand group, extender group, parallel linking unit, PEG group, release complex group, and drug group), the synthesis of ligand-drug conjugates drug formulations including additional drug groups can be easily implemented using the disclosure presented in this application. The inclusion of additional branching groups and/or drug attachment groups allows multiple drug groups to be attached to the PEG group. Additional -X-D groups are added via branching or drug addition groups.

In one group of embodiments, such LDC compounds capable of releasing free drug are represented by formulas (I), (II) or (III):

- 17 046498 drug-linker drug-linker drug-linker (III) or a pharmaceutically acceptable salt of such a compound, where

L represents a ligand group;

D represents a drug group;

PEG is a polyethylene glycol group;

Z represents an expansion group;

X represents a releasable complex group;

L ^P represents a parallel connecting link;

A represents an optional branching group;

AD represents the drug addition group;

the index p is an integer having a value from 1 to 14, preferably from 2 to 12 (preferably from 6 to 14, from 6 to 12, from 8 to 14 or from 8 to 12);

index t is an integer having a value from 0 to 8 and preferably has a value of 0, 1, 2 or 3;

index m is an integer ranging from 1 to 4; and preferably has a value of 1 or 2; and the index s has the value 0 or 1, with the proviso that when s has the value 0, m has the value 1, and when s has the value 1, m has the value 2, 3, or 4.

In another group of embodiments, formulas I, II and III do not represent individual LDC compounds, but rather an LDC composition (ie, a composition comprising a population of individual LDC compounds). In such embodiments, p represents the average number of drug-linker groups per ligand in the composition. In such embodiments, p is typically not an integer value and may range from 1 to about 14, preferably from about 2 to about 12 (preferably from about 6 to about 14, from about 6 to about 12, from about 8 to about 14 or from about 8 to about 12). Other variables (eg L, Z, A, L ^P , PEG, X, D, AD, s, m and t) have the meaning defined above.

In another group of embodiments, the LDC composition includes a population of LDC compounds, wherein the individual LDC compounds are represented by formula I, II or II, wherein, for each individual LDC compound, p is independently selected from an integer having a value from 1 to 14, preferably from 2 to 12 (preferably 6 to 14, 6 to 12, 8 to 14, or 8 to about 12), and the average number of drug-linker groups per ligand in the composition is from 1 to about 14, preferably from about 2 to about 12 (preferably from about 6 to about 14, from about 6 to about 12, from about 8 to about 14, or from about 8 to about 12).

In some aspects, from 1 to 32 or from 2 to 32 (preferably from 6 to 32 or 8 to 32) drug groups are attached to each ligand group. The population of drug-ligand conjugates may contain an average of 1 to 32, or about 2 to 32 (preferably

- 18 046498 but from about 6 to 32, or about 8 to 32) drug groups per ligand.

Specific embodiments of formulas I, II and III include those in which:

1) m has the value 1 and s has the value 0;

2) m has the value 2, 3 or 4 and s has the value 1;

3) m has the value 2 and s has the value 1;

4) m has the value 1; s has the value 0; and p is an integer having a value from 2 to 12, 4 to 12, 8 to 14, or 8 to 12, for a ligand-drug conjugate compound, or p is a number having a value from about 2 to about 12, about 4 to about 12, about 8 to about 14, or about 8 to about 12, for a ligand-drug conjugate composition;

5) m has the value 2, 3 or 4; s has the value 1; and p is an integer having a value of from about 2 to about 12, from about 4 to about 12, from about 8 to about 14, or from about 8 to about 12 for a ligand-drug conjugate compound, or p is a number , having a value of from about 2 to about 12, from about 4 to about 12, from about 8 to about 14, or from about 8 to about 12, for a ligand-drug conjugate composition;

6) m has the value 2; s has a value of 1 and p is an integer having a value of 2 to 12, 4 to 12, 6 to 14, 6 to 12, 8 to 14, or about 8 to about 12 for a Ligand-Drug conjugate compound , or p is a number having a value of about 2 to about 12, from about 4 to about 12, from about 6 to about 14, from about 6 to about 12, from about 8 to about 14, or from about 8 to about 12 for ligand-drug conjugate compositions;

7) m has the value 2; s has the value 1; and p has a value of 8;

8) m has the value 1; s has the value 0; and p has a value of 8;

9) any of the embodiments described in paragraphs 1-8 of this subsection, where t has the value 0;

10) any of the embodiments described in paragraphs 1-8 of this subsection, where t has a value of 1-8;

11) any of the embodiments described in paragraphs 1-8 of this subsection, where t has the value 1;

12) any of the embodiments described in paragraphs 1-8 of this subsection, where t has a value of 2;

13) any of the embodiments described in paragraphs 1-8 of this subsection, where t has a value of 3;

14) any of the embodiments described in paragraphs 1-8 of this subsection, where t has a value of 4;

15) any of the embodiments described in paragraphs 1-8 of this subsection, where t has a value of 5;

16) any of the embodiments described in paragraphs 1-8 of this subsection, where t has a value of 6;

17) any of the embodiments described in paragraphs 1-8 of this subsection, where t has a value of 7;

18) any of the embodiments described in paragraphs 1-8 of this subsection, where t has a value of 8;

19) any of the embodiments described in claims 1 to 18 of this subsection, wherein from 1 to 32 or about 2 to 32 drug groups are attached to the ligand group;

20) any of the embodiments described in claims 1 to 18 of this subsection, wherein from 6 to 32 or about 8 to 32 drug groups are attached to the ligand group; And

21) any of the embodiments described in paragraphs 1-20 of this subsection, where L ^P represents a natural or non-natural amino acid, amino alcohol, aminoaldehyde or polyamine.

In any of the selected embodiments for LDC compounds provided herein where p is present, including the embodiments described above, p may be an integer having a value of 1 to 14, 2 to 14, 2 to 10 , from 4 to 12, from 6 to 14, from 6 to 12, from 8 to 12 or from 8 to 10. The index p can have the value 1 or 2 or 3 or 4 or 5 or 6 or 7 or 8 or 9 or 10 or 11 or 12 or 13 or 14.

In any of the selected embodiments for LDC compositions provided herein where a p value is present, including the embodiments described above, p has a value of 1 to about 14, about 2 to about 14, about 2 to about 10, from about 4 to about 12, from about 6 to about 14, from about 6 to about 12, from about 8 to about 12, or from about 8 to about 10. The index p may have a value of 1 or about 1, or 2 or about 2 , or 3 or about 3, or 4 or about 4, or 5 or about 5, or 6 or about 6, or 7 or about 7, or 8 or about 8, or 9 or about 9, or 10 or about 10, or 11 or about 11, or 12 or about 12, or 13 or about 13, or 14 or about 14. Other variables (for example, L, Z, A, L ^P , PEG, X, D, AD, s, m and t ) have the meaning defined above.

Specific embodiments of Formulas I, II, and III include Formulas Ia, Ib, IIa, IIb, IIb, IIIa, and

- 19 046498

IIIb, presented below.

or a pharmaceutically acceptable salt of such a compound, wherein,

L represents a ligand group;

D represents a drug group;

PEG is a polyethylene glycol group;

Z represents an expansion group;

X represents a releasable complex group;

L ^P represents a parallel connecting link;

A represents an optional branching group and

AD represents the drug addition group;

subscript p is an integer having a value of 1 to 14, preferably 2 to 12 (preferably 6 to 14, 6 to 12, 8 to 14, or 8 to 12) for a ligand-drug conjugate compound means, or subscript p means a number having a value from 1 to about 14, preferably from about 2 to about 12 (preferably from about 6 to about 14, from about 6 to about 12, from about 8 to about 14, or from about 8 up to about 12) for a ligand-drug conjugate composition; and the subscript t is an integer having a value from 0 to 8; and preferably has a value of 0, 1, 2 or 3.

Specific embodiments of formulas Ia, Ib, IIa, IIb, IIb, IIc, IIIa and IIIb include those in which:

1) t has the value 0;

2) t has a value from 1 to 8;

3) t has the value 1;

4) t has the value 2;

5) t has a value of 3;

6) t has a value of 4;

7) t has a value of 5;

8) t has a value of 7;

9) t has a value of 8;

10) any of the embodiments described in paragraphs 1 to 10 of this subsection, wherein 1 to 32, about 2 to 32, 6 to 32, or about 8 to 32 drug groups are attached to a Ligand Group; And

11) any of the embodiments described in paragraphs 1-11 of this subsection, where L ^P represents a natural or non-natural amino acid, amino alcohol, amino aldehyde or polyamine.

Embodiments of formulas Ia, Ib, IIa, IIb, IIb, IIc, IIIa and IIIb for LDC compositions include

- 20 046498 such in which p is a number having a value from 6 to about 12; from about 8 to about 12, and from about 8 to about 10. For these compositions, the subscript p may be 6 or about 6, or 7 or about 7, or 8 or about 8, or 9 or about 9, or 10 or about 10, or 11, or about 11, or 12, or about 12, or 13, or about 13, or 14, or about 14. In any of these embodiments, t may be from 0 to 8, from 1 to 8, or 0 , 1, 2, 3, 4, 5, 6, 7 or 8.

Embodiments of formulas Ia, Ib, IIa, IIb, IIb, IIc, IIIa and IIIb for an LDC compound include those in which p is an integer ranging from 6 to 12; 8 to 12 and 8 to 10. The subscript p may have a value of 6, 7, 8, 9, 10, 11, 12, 13, or 14. In any of these embodiments, t may have a value of 0 to 8. from 1 to 8, or 0, 1, 2, 3, 4, 5, 6, 7 or 8.

Drug-linker compounds

Below are examples of drug-linker compounds containing at least 1 drug per PEG group, having formulas IV, V, VI

P.E.G.

or a pharmaceutically acceptable salt of such a compound, wherein

D represents a drug group;

PEG is a polyethylene glycol group;

Z' represents an extending group capable of forming a covalent attachment to a ligand group;

X represents a releasable complex group;

L ^P represents a parallel connecting link;

A represents an optional branch;

AD represents the drug addition group;

subscript t represents an integer ranging from 0 to 8; and preferably has a value of 0, 1, 2 or 3;

subscript m is an integer ranging from 1 to 4; and preferably has a value of 1 or 2;

subscript s has the value 0 or 1, provided that when s has the value 0, m has the value 1, and when s has the value 1, m has the value 2, 3, or 4.

Certain embodiments of formulas IV, V and VI include those in which

1) m has the value 1 and s has the value 0;

2) m has a value from 2 to 4, and s has a value of 1;

3) m has the value 2 and s has the value 1;

4) any of the embodiments described in paragraphs 1-3 of this subsection, where t has the value 0;

5) any of the embodiments described in paragraphs 1-3 of this subsection, where t has the value 1;

6) any of the embodiments described in paragraphs 1-3 of this subsection, where t has a value of 2; And

7) any of the embodiments described in paragraphs 1-6 of this subsection, where L ^P represents a natural or non-natural amino acid, amino alcohol, amino aldehyde or polyamine.

Certain embodiments of formulas IV, V and VI include the following formulas:

- 21 046498

or a pharmaceutically acceptable salt of such a compound, wherein D represents a drug group; PEG is a polyethylene glycol group;

Z' represents an extending group capable of forming a covalent attachment to a ligand group;

X represents a releasable complex group;

L ^P represents a parallel connecting link;

A represents an optional branch;

AD represents the drug addition group and the subscript t is an integer having a value from 0 to 8; and preferably has a value of 0, 1, 2 or 3.

Linker intermediates

Below are examples of linker intermediates comprising at least one drug per PEG group having formulas VII, VIII or IX

or a pharmaceutically acceptable salt of such a compound, wherein PEG represents a polyethylene glycol group;

Z' represents an extending group capable of forming a covalent attachment to a ligand group;

A' is a branching group capable of forming a covalent attachment to two to four X-D groups, preferably two X-D groups;

A represents an optional branching group;

AD' represents a drug attachment group capable of forming a covalent attachment to the -X-D group;

- 22 046498

LP is a parallel connecting link;

LP is a parallel connecting unit capable of forming a covalent attachment to -X-D;

subscript t is an integer ranging from 0 to 8, and is preferably 0, 1, 2 or 3;

subscript m is an integer ranging from 1 to 4; and preferably has a value of 1 or 2;

subscript s has the value 0 or 1, provided that when s has the value 0, m has the value 1, and when s has the value 1, m has the value 2, 3, or 4; and where -X-D represents a releasable complex group attached to a drug group.

Specific embodiments of formulas VIII or IX include the following:

or a pharmaceutically acceptable salt of such a compound, wherein PEG represents a polyethylene glycol group;

Z' represents an extending group capable of forming a covalent attachment to a ligand group;

A represents a branching group;

AD' represents a drug attachment group capable of forming a covalent attachment to the -X-D group;

L ^P represents a parallel connecting link;

LP is a parallel connecting unit capable of forming a covalent attachment to -X-D; and subscript t is an integer having a value from 0 to 8; and preferably has a value of 0, 1, 2 or 3; and where -X-D represents a releasable complex group attached to a drug group.

In linker intermediates and formulas VII, VIII, XI, VIIIa, VIIIb, VIIIc, VIIId, IXa and IXb, an extension group can be conjugated to a ligand group (eg, antibody) to form ligand-linker intermediates that provide 1 to 14 linkers attached to each ligand group. Exemplary embodiments are shown below, where p has a value from 1 to 14 and all other variables have the values defined herein for linker intermediates. The following are examples of ligand-linker compounds and compositions comprising these compounds (i.e., ligand-linker compositions) that have the structures represented by formulas X, XI, XII

- 23 046498

linker (XII) or a pharmaceutically acceptable salt of such a compound, where

L represents a ligand group;

PEG is a polyethylene glycol group;

Z- represents an extension group;

-X-D is a releasable complex group attached to a drug group;

L ^P represents a parallel connecting link;

L ^P is a parallel connecting unit capable of forming a covalent attachment to -XD;

A is a branching group capable of forming a covalent attachment to two to four X-D groups, preferably two X-D groups;

A represents an optional branching group;

AD' represents a drug attachment group capable of forming a covalent attachment to the X-D group;

subscript p is an integer having a value from 1 to 14, preferably from 2 to 12 (preferably from 6 to about 14, from about 6 to about 12, from about 8 to about 14, or from about 8 to about 12) for The compound linker ligand, or subscript p, is a number ranging from 1 to about 14, preferably about 2 to about 12 (preferably about 6 to about 14, about 6 to about 12, about 8 to about 14, or about 8 to about 12) for a ligand-linker composition;

subscript t has a value from 0 to 8; and preferably has a value of 0, 1, 2 or 3;

subscript m is an integer ranging from 1 to 4; and preferably has a value of 1 or 2; and the subscript s has the value 0 or 1, with the proviso that when s has the value 0, m has the value 1, and when s has the value 1, m has the value 2, 3, or 4.

Certain embodiments of formulas XI and XII include the following formulas.

- 24 046498

or a pharmaceutically acceptable salt of such compound where

L represents a ligand group;

PEG is a polyethylene glycol group;

Z- represents an extension group;

L ^P represents a parallel connecting link;

L ^P is a parallel connecting unit capable of forming a covalent attachment to -XD;

A represents a branching group;

AD' is a drug attachment group capable of forming a covalent attachment to the X-D group;

subscript p is an integer having a value of 1 to 14, preferably 2 to 12 (preferably 6 to 14, 6 to 12, 8 to 14, or 8 to 12) for a ligand-linker compound, or subscript subscript p means a number having a value from 1 to about 14, preferably from about 2 to about 12 (preferably from about 6 to about 14, from about 6 to about 12, from about 8 to about 14, or from about 8 to about 12) for ligand-linker compositions; and the subscript t has a value from 0 to 8; and where -X-D is a releasable complex group attached to a drug group.

Components of the group

Common to the ligand-drug conjugates and intermediates described herein is the placement of the PEG moiety in a parallel orientation with its drug moiety to influence the pharmacokinetics of the resulting LDC. The placement of the PEG group is carried out using a parallel connecting link. The parallel connecting link serves to connect the ligand to the polyethylene glycol group and the drug group so that the PEG and drug groups are in a parallel configuration, which ensures that the ligand, PEG and drug groups are arranged in a branched configuration.

Accordingly, the parallel connecting link can be considered a framework containing attachment sites for components of ligand-drug conjugates and intermediates for their preparation.

To act as a parallel connecting link, the LP link is attached through three attachment sites in the linker. One of the connection sections connects LP Link to the PEG Group. The second attachment site attaches the LP Unit to a releasable complex group (in some cases via a branching group A or a drug attachment group AD). The third attachment site attaches the LP unit to the extension group (in some cases through the drug attachment group AD and/or the branching group A). A parallel connecting unit is a unit that is different from the PEG group and is attached to it through the PEG attachment component of the PEG group. That is, the parallel connecting link is not a subunit of the PEG group.

With respect to ligand-drug conjugates and their intermediates containing more than one drug per PEG group, the attachment of a parallel linking unit to the releasable complex group may occur via a branching

- 25 046498 group or group of addition of a medicinal product. Attachment of the parallel connecting link to the extension group may occur through the drug attachment group AD and/or optionally through an additional branching group. In all of these embodiments, the L ^P unit can be considered a tri-functional chemical group that is capable of covalently linking together three spaced apart chemical groups. As will be appreciated, for individual L ^P intermediates the unit is represented as L ^P and is not yet attached to the drug via a drug releasing group, but contains an optionally protected functional group for attachment to the drug (e.g. via a drug releasing group .) It should also be understood that the term trifunctional is used to indicate three sites of attachment, and not the number of functional groups present in the L ^P or L ^P unit.

The parallel connecting unit can be formed from one or more (typically 1 to 5, or 1 to 4, or 1 to 3, or 1 or 2) naturally occurring or non-naturally occurring amino acids, amino alcohols, amino aldehydes, or polyamines.

It should be understood that when referring to a natural or non-naturally occurring amino acid, amino alcohol, aminoaldehyde or polyamine present in a conjugate or intermediates of the present invention (whether they are part of the L ^P Unit or other component of the conjugates or intermediates compounds described herein), the amino acid, amino alcohol, amino aldehyde or polyamines exist in a residual form, also referred to herein as a combined form. For example, in embodiments where the parallel connecting unit is two amino acids, the two amino acids will exist as residues with a peptide bond between them. In embodiments where the parallel link unit consists of an amino alcohol, the amino alcohol will exist as a moiety where, for example, its amino group is linked to another parallel link unit moiety or other conjugate component via a carbonyl-containing functional group of such other moiety/component, where In this case, its hydroxyl group is ether-linked to, or linked through a carbonyl-containing functional group, another parallel linking moiety or other conjugate component. In embodiments where the parallel link unit consists of an amino aldehyde, the amino aldehyde will exist as a moiety where, for example, its amino group is linked to another parallel link unit moiety or other conjugate component via a carbonyl-containing functional group of such other moiety/component, wherein its aldehyde functional group is converted to an imino functional group, or through subsequent reduction to provide a nitrogen-carbon bond upon binding to the amino group of another residue of the parallel connecting unit or other component of the conjugate. An amino alcohol or amino aldehyde can be formed from a natural or non-naturally occurring amino acid by reducing its carboxylic acid functionality to an aldehyde or hydroxyl functionality.

When the parallel link moiety is a branching moiety for that unit, it will be understood that the moiety will contain a third functional group to which another parallel link moiety, an -X-D group or a PEG group or other linker group component, binds. For example, an amino acid or other amine-containing acidic moiety of a parallel linking unit may contain or may be substituted with a functionalized side chain to provide the required three attachment points required for the branching moiety. For example, serine contains three functional groups, i.e. acid, amino and hydroxyl functional groups, and can be considered as a combined amino acid and amino alcohol residue for the purpose of its inclusion in a parallel connecting link. Tyrosine also contains a hydroxyl group, in this case on its phenolic side chain, and can also be treated similarly to serine for the purposes of its inclusion as a branching residue in a parallel connecting unit.

In another example, when the branching residue of the parallel linking unit is a cysteine, its amino and carboxylic acid group will exist in residual form, just as discussed above for amino acids or amine-containing acids, to provide two of the three required attachment points for the branching residue, while how its thiol group will exist in residual form when it is linked to an -XD group or PEG group or other component of the linker group as a disulfide, or as a sulfur-carbon bond, for example, when the thiol functional group reacts with the maleimide-containing group of the component Linker Group. In some cases, the residual thiol group is in its oxidized form (i.e., -S(=O)- or -S(=O) ₂ -) when linked to another parallel linking moiety or to another linker group component . In yet another example, the alpha amino and carboxylic acid group of lysine will exist in residual form to provide two of the three necessary attachment points required for the branching residue of the parallel link unit, while its epsilon amino group in its residual form provides the third attachment point. Ghisti

- 26 046498 Dine can also be considered as an amino acid with two amino groups, where the second amino group is the NH of the imidazole-containing side chain.

In another example, when the branching moiety of the parallel link unit is aspartic or glutamic acid, the alpha amino and C-terminal carboxylic acid groups of the amino acid in their residual forms provide two of the three necessary attachment points required for the branching moiety of the parallel linking unit, while its beta or gamma carboxylic acid group in its residual form provides a third point of attachment. In cases where a naturally occurring amino acid is described as a parallel linkage residue but does not inherently contain a functionalized amino acid side chain but is required to be a branching residue, it will be understood that the amino acid structure is modified to contain an additional functional group in addition to its amino and carboxylic acid functional groups when in residual form, to provide the required third point of attachment. For example, an amino acid containing an aliphatic side chain may be substituted at a carbon of that side chain with a hydroxyl, amino, aldehyde, thiol, carboxylic acid group, or other functional group, or another group (e.g., aryl or arylalkyl) substituted with any of these functional groups, with by obtaining a non-naturally occurring amino acid containing the required three attachment points. Such non-naturally occurring amino acids are incorporated into the parallel linking unit as described above for amino acids and the residual forms of the introduced functional groups.

Likewise, when an amino aldehyde or amino alcohol is incorporated into a parallel linking unit as a branching moiety, the amino aldehyde or amino alcohol will contain a third functional group to provide, along with its amino and aldehyde functional groups, the required three points of attachment. In these cases, the aminoaldehyde or amino alcohol may correspond in structure to a naturally occurring amino acid that contains a functionalized side chain, or to a non-naturally occurring amino acid containing a functional group that has been introduced into the naturally occurring amino acid side chain, as described above, in which the naturally occurring or non-naturally occurring carboxylic acid a naturally occurring amino acid is reduced to a hydroxy or aldehyde functional group.

An amino acid may be an alpha, beta or gamma amino acid or other amine-containing acidic compound and may be present as its D or L isomer if it contains a chiral carbon to which a side chain of a natural or non-naturally occurring amino acid is linked. When the parallel linker consists of more than one natural or non-naturally occurring amino acid, amino alcohol, amino aldehyde or polyamines, the amino acids, amino alcohols, amino aldehydes, polyamines or combinations thereof are linked together through covalent bonds to form the parallel linker.

The amino acid, amino alcohol or amino aldehyde may not be naturally occurring and may be modified to have a functionalized side chain for attachment to components of conjugates or intermediates (as described above for the branching moiety of the parallel linker), as appropriate. Examples of functionalized amino acids, amino alcohols or amino aldehydes include, for example, azido or alkyne functionalized amino acids, amino alcohols or amino aldehydes (for example, an amino acid, amino alcohol or amino aldehyde modified to contain an azide group or an alkyne group for addition using click chemistry). Methods for independently activating and interacting functional groups present on an amino acid—for example, an amine moiety, a carboxylic acid moiety, and a side chain (whether, for example, an amino group, a hydroxyl group, another carboxylic acid, a thiol, an azide, or an alkyne)—are well known in the art. .

The parallel linking unit may include 1 or more (typically 1 to 5, or 1 to 4, or 1 to 3, or 1 or 2) amino acids, optionally substituted C _1-20 heteroalkylenes (preferably optionally substituted C _1-12 heteroalkylene ), optionally substituted _C3-8 heterocyclo, optionally substituted _C6-14 arylenes, optionally substituted C3-C8 carbocyclo, or combinations thereof. In some aspects, the Parallel Connecting Unit includes no more than 2 or no more than one optionally substituted C _1-20 heteroalkylene, optionally substituted C ₃₈ heterocyclo, optionally substituted C _6-14 arylene, or optionally substituted C ₃ -C ₈ carbocyclo. Optional substituents include (=O), -X, -R, -OR, -SR, -NR ₂ , -NR ₃ , =NR, -CX ₃ , -CN, -oCn, -SCN, -N=C=O , -NCS, -NO, -NO2, =N2, -N3, -NRC(=O)R, -C(=O)R, -C(=O)NR2, -SO3-, -SO3H, -S( =O)2R, -OS(=O)2OR, -S(=O)2NR, -S(=O)R, -OP (=O)(OR)2, - P(=O)(OR)2 , -PO=3, -PO3H2, -ASO2H2, -C(=O)R, -C(=O)X, -C(=S)R, -CO2R, -CO2-, -C(=S)OR , C(=O)SR, -C(=S)SR, -C(=O)NR2, -C(=S)NR2 or -C(=NR)NR2, where each X is independently a halogen: -F , -Cl, -Br or -I; and each R independently represents -H, -C1 CCylkyl. -C ₆ C ₂₀ aryl, -C ₃ ^ heterocycle, protecting group or prodrug group. Preferred optional substituents are (=O), -X, -R, -OR, -SR and -NR2.

- 27 046498

The parallel connecting link can be a linear or branched chain and can be represented by formula A

I LLLR

I

Formula A, where

AA ¹ is an LP subunit independently selected from an amino acid, an optionally substituted C _1-20 heteroalkylene (preferably an optionally substituted C _1-12 heteroalkylene), an optionally substituted C _3-8 heterocyclo, an optionally substituted C _6-14 arylene, or an optionally substituted C3- C8 carbocyclo;

and subscript and independently selected from values from 0 to 4; and the wavy line indicates the sites of covalent attachment in the ligand-drug conjugate or intermediate thereof. The optionally substituted heteroalkylene, heterocycle, arylene or carbocyclo must contain functional groups for attachments between subunits and in the ligand-drug conjugate or intermediates thereof.

In some aspects, in at least one instance, AA ¹ is an amino acid. The subscript and may have a value of 0, 1, 2, 3, or 4. In some aspects, AA ¹ is an amino acid, and and has a value of 0. In some aspects, the parallel connecting unit includes no more than 2 optionally substituted C _1-20 heteroalkylenes , optionally substituted C _3-8 heterocyclo, optionally substituted C _6-14 arylenes or optionally substituted C ₃ -C ₈ carbocyclo. In some aspects, when the parallel connecting unit has Formula A, the parallel connecting unit includes no more than 1 optionally substituted C _1-20 heteroalkylene, optionally substituted C _3-8 heterocyclo, optionally substituted C _6-14 arylene, or optionally substituted _{C 3} -C ₈ carbocyclo.

The parallel linker or amino acid subunit thereof may be an alpha, beta or gamma amino acid, which may be naturally occurring or non-naturally occurring. The amino acid may be a D or L isomer. Coupling in a parallel link or with other components of the conjugate (or linker) can be accomplished, for example, through amino, carboxy or other functional groups. Methods for independent activation and interaction of functional groups are well known in the art.

The parallel linking unit or amino acid subunit thereof may be independently selected from the D or L isomer of the thiol-containing amino acid. The thiol-containing amino acid may be, for example, cysteine, homocysteine or penicillamine.

The parallel connecting unit or amino acid subunit thereof may be independently selected from the group consisting of L- or D-isomers of the following amino acids: alanine (including β-alanine), arginine, aspartic acid, asparagine, cysteine, histidine, glycine, glutamic acid, glutamine phenylalanine, lysine, leucine, methionine, serine, tyrosine, threonine, tryptophan, proline, ornithine, penicillamine, B-alanine, aminoalkynoic acid, aminoalkandioic acid, heterocyclocarboxylic acid, citrulline, statin, diaminoalkanoic acid and their derivatives.

Preferred amino acids include cysteine, homocysteine, penicillamine, ornithine, lysine, serine, threonine, glutamine, alanine, aspartic acid, glutamic acid, selenocysteine, proline, glycine, isoleucine, leucine, methionine, valine and alanine.

Examples of L ^P or its AA ¹ subunits include

- 28 046498

ch ₂ ch ₂ coo-£- ₅ *-CH ₂ *—(CH ₂ ) ₄ NHC(=N-NH)CH ₃ , where R ¹¹⁰ represents *-ch ₂ o-£·, *—(CH ₂ ) ₃ NHC(=NH)NH ^- , *--(CH ₃ ) ₄ NHC(=NO)CH ₃ , *— ССОСНз ' *-(CH ₂ ) ₃ NH-^*---(CH ₂ ) ₃ NHCONH-|- , *—ch ₂ conh-|— , *---(CH ₂ ) ₃ NHC(=N-NH)CH ₃ , *—CH ₂ COO-|*---(CH ₂ ) ₃ NHC (=NO)CH ₃ , *--CH ₂ CH ₂ CH(OH)CH ₂ NH—

--CH ₂ CH ₂ OCH ₂ CH ₂ NH—£>*—СНгСНгСОН-^' *----(CH ₂ ) ₃ NHCH=N-NH-|- _j *—(CH ₂ ) ₄ NHC(=NH )NH-S, (CH ₂ )i. ₄ NH-?*---(CH ₂ ) ₃ NHCH=NO—cj- , *--(CH ₂ ) ₄ NHCONH-^*—(CH ₂ )1. ₄ S-^- , *— (C(CH ₃ )(CH ₃ )s4 *—(C(CH ₃ )(CH ₃ )NH-<-

*—CH ₂

or *—CH ₂

R is independently selected from hydrogen, p-hydroxybenzyl, methyl, isopropyl, isobutyl, sec-butyl, -CH2OH, -CH(OH)CH3, -CH2CH2SCH3, -CH2CONH2, -CH2COOH, -CH2CH2CONH2, -CH2CH2COOH, (CH2)3NHC(=NH )NH2, -(CH2)3NH2,

-(CH2)4NH2, -(CH2)4NHCOCH3,

-(CH2)3ННСОНЗ, -(CH2)4NHCHO,

-(CH2)3NHCHO, (CH2)4NHC(=NH)NH2, (CH2)3NHCONH2, -(CH2)4NHCONH2,

- 29 046498

-CH ₂ CH2CH(OH)CH ₂ NH2, 2-pyridylmethyl-, 3-pyridylmethyl-, 4-pyridylmethyl-,

where the asterisk indicates attachment to the carbon labeled x;

R ¹⁰⁰ is independently selected from hydrogen or -C ₁ -C ₃ alkyl (preferably hydrogen or CH ₃ ),

R ¹³ is independently selected from the group consisting of the following: -C1-C6 alkylene-, -C ₃ -C ₈ carbocyclo-, -arylene-, -C ₁ -C _1o heteroalkylene-, -C ₃ -C ₈ heterocyclo-, - ^-^alkylene-arylene-, -arylene-C ₁ -C ₁₀ alkylene-, -^-C^alkylene- (C ₃ -C ₈ carbocyclo)-, -(C ₃ -C ₈ carbocyclo) -^-C^ alkylene-, -^-C^alkylene-^^heterocyclo)- and -(C ₃ -C ₈ heterocyclo^^-C^ alkylene- (preferably -CH2-CH2-);

Y represents - C(R ¹⁰⁰ ) - or -

Y' is -C(=O)-, -O-, -S-, -NH- or -N(CH ₃ )- and the subscripts p, q and d are integers independently selected from the values 0- 5; and the wavy line indicates a covalent addition to the compound, a hydrogen, OH, or C _1-3 unsubstituted alkyl group, provided that at least one of the wavy lines indicates a covalent addition to the compound. In some aspects, all wavy lines indicate covalent attachment in the compound (eg, when the LP does not include any subunits).

In one group of embodiments, LP is a heterocyclic ring containing functional groups that can independently form covalent bonds with these components (eg, a triazole heterocyclic ring formed from cyanuric acid chloride). In another group of embodiments, LP is an alkane containing attached functional groups as described above. In some other embodiments, LP may be a nitrogen atom.

In some embodiments, -LP-, when assembled, has the formula shown below

wherein the wavy line indicates the sites of attachment in a ligand-drug conjugate or intermediate thereof (eg, PEG to -X (directly or indirectly through A or AD) and to Z (directly or indirectly through A or AD), and where R ¹¹⁰ represents yourself

- 30 046498

*-sn ₂ o-<;*—СНСНз *—ch ₂ conh-|— *—sn ₂ coo-|*—ch ₂ ch ₂ conh-^*—(ch ₂ ) ₄ nhc(=nh)nh-;

, *—(CH ₂ ) ₃ NHC(=NH)NH ' *-(CH ₂ ) ₃ NH-^- , = '---(CH ₂ ) ₃ NHC(=N-NH)CH ₃ , = *- --(CH ₂ ) ₃ NHC(=NO)CH ₃ ' *----(CH ₂ ) ₃ NHCH=N-NH-^- _; , *—(CH ₂ )1. ₄ NH-|- = *---(CH ₃ ) ₄ NHC(=NO)CH ₃ *---(CH ₂ ) ₃ NHCONH-|- ₅ '---CH ₂ CH ₂ CH(OH)CH ₂ NH-^*—CH ₂ CH ₂ CH(O)CH ₂ NH ₂ , *---(CH ₂ ) ₃ NHCH=NO-^- ₅ *--(CH ₂ ) ₄ NHCONH-£-

where the asterisk indicates attachment to the carbon labeled x and the wavy line indicates one of three attachment sites;

R ¹⁰⁰ is independently selected from hydrogen or -C1-C3 alkyl, preferably hydrogen or CH3,

Y is independently selected from N or CH,

Y' is independently selected from NH, O or S and the subscript c is an integer independently selected from 1-10, and preferably 1, 2 or 3.

In preferred embodiments, R ¹¹⁰ is other than *—CH2CH ₂ CH(O)CH ₂ NH2

I

The parallel connecting unit or amino acid subunit thereof may have the formula below

where the subscript n represents an integer ranging from 1 to 4;

XP is selected from the group consisting of -O-, -NR-, -S-, -S(=O)-, -C(=O)- or -C2-C8 heterocyclo-; And

R ¹ and R ² are independently selected from the group consisting of -H, -C _1-3 alkyl, -phenyl or -C 2 -C 5 heterocycle (preferably H or C _1-3 alkyl), where the wavy line indicates covalent attachment in the compound.

In some embodiments, XP is provided by a side chain of a natural or non-naturally occurring amino acid.

Each parallel connecting unit or subunit thereof may be independently selected from the D or L isomer of lysine, glutamic acid, aspartic acid, cysteine, penicillamine, serine or threonine.

Each parallel connecting unit or subunit thereof may be independently selected from the D or L isomer of lysine, glutamic acid, aspartic acid, cysteines or penicillamines.

Each parallel connecting link or subunit thereof may be independently selected from the group consisting of the following amino acids: arginine, aspartic acid, asparagine, histidine, glutamic acid, glutamine, lysine, serine, tyrosine, threonine, tryptophan, ornithine, penicillamine, aminoalkynoic acid, aminoalkandioic acid, heterocyclocarboxylic acid, citrulline, statin, diaminoalkanoic acid and their derivatives.

Each parallel connecting link or its subunit can be independently selected

- 31 046498 from the group consisting of the following L-isomers of these natural amino acids: arginine, aspartic acid, asparagine, histidine, glutamic acid, glutamine, lysine, cysteine, penicillamine, serine, tyrosine, threonine and tryptophan.

Each parallel linking unit or subunit thereof may be independently selected from the group consisting of the following D-isomers of these natural amino acids: arginine, aspartic acid, asparagine, histidine, glutamic acid, glutamine, phenylalanine, lysine, cysteine, penicillamine serine, tyrosine, threonine , and tryptophan.

Each parallel connecting unit or subunit thereof may be independently selected from the D or L isomer of the thiol-containing amino acid. The thiol-containing amino acid may be, for example, cysteine, homocysteine or penicillamine.

Each parallel connecting unit or subunit thereof may be independently selected from the group consisting of L- or D-isomers of the following amino acids: alanine (including β-alanine), arginine, aspartic acid, asparagine, cysteine, histidine, glycine, glutamic acid, glutamine phenylalanine, lysine, leucine, methionine, serine, tyrosine, threonine, tryptophan, proline, ornithine, penicillamine, B-alanine, aminoalkynoic acid, aminoalkandioic acid, heterocyclocarboxylic acid, citrulline, statin, diaminoalkanoic acid and their derivatives.

Preferred amino acids include cysteine, homocysteine, penicillamine, ornithine, lysine, serine, threonine, glutamine, alanine, aspartic acid, glutamic acid, selenocysteine, proline, glycine, isoleucine, leucine, methionine and valine.

Each parallel connecting unit or subunit thereof may be independently selected from the group consisting of alanine derivatives, provided that a suitable number of functional units is present. Illustrative examples of alanine derivatives include, but are not limited to: dehydro-alanine, 4-thiazolylalanine, 2-pyridylalanine, 3-pyridylalanine, 4-pyridylalanine, β-(1-naphthyl)alanine, β-(2-naphthyl)alanine, α -aminobutyric acid, β-chloroalanine, β-cyanoalanine, β-cyclopentyl-alanine, β-cyclohexyl-alanine, β-iodo-alanine, β-cyclopentenyl-alanine, e-tBu-alanine, β-cyclopropyl-alanine, β-diphenyl-alanine, β-fluoro-alanine, β-piperazinyl-alanine, where the piperazine ring is protected or unprotected, β-(2-quinolyl)alanine, β-(1,2,4-triazol-1-yl)alanine, β-ureidoalanine, H-β-(3-benzothienyl)-Ala-OH and H-β-(2-thienyl)-Ala-OH.

Each Parallel Connecting Unit or subunit thereof may be independently selected from the group consisting of arginine and arginine derivatives. Illustrative examples of arginine and its derivatives include, but are not limited to: arginine (Arg), N-alkyl-arginine, H-Arg(Me)-OH, HArg(NH ₂ )-OH, H-Arg(ΝΟ ₂ )-ΟΗ , H-Arg(Ac) ₂ -OH, H-Arg (Me) ₂ -OH (asymmetric), H-Arg(Me) ₂ -OH (symmetric), 2-amino-4-(2'-hydroxyguanidino)- butyric acid (Ν-ω-hydroxy-nor-arginine) and homoarginine.

Each parallel connecting unit or subunit thereof may be independently selected from the group consisting of aspartic acid and its derivatives. Illustrative examples of aspartic acid and its derivatives include, but are not limited to: aspartic acid (Asp), N-alkyl-aspartic acid and H-Asp(OtBu)-OH.

Each parallel connecting unit or subunit thereof may be independently selected from the group consisting of asparagine and its derivatives. Illustrative examples of asparagine and its derivatives include, but are not limited to: asparagine (Asn), N-alkyl-asparagine and isoasparagine (H-Asp-NH2).

Each parallel connecting unit or subunit thereof may be independently selected from the group consisting of cysteine and its derivatives. Illustrative examples of cysteine (Cys) and its derivatives (not containing a free SH group) include, but are not limited to: Cys(StBu), H-Cys(Acm)-OH, H-Cys(Trt)-OH, H-Cys( StBu)-OH, H-Cys(Bzl)-OH, H-Cys(S-Et)-OH, H-Cys(SO ₃ H)-OH, HCys(aminoethyl)-OH, H-Cys(carbamoyl)- OH, H-Cys(S-phenyl)-OH, H-Cys(Boc)-OH and H-Cys(hydroxyethyl)-OH.

Each Parallel Link Unit or subunit thereof may be independently selected from the group consisting of histidine and its derivatives. Illustrative examples of histidine and its derivatives include, but are not limited to: histidine (His), N-alkyl-histidine, H-His(Boc)-OH, HHis(Bzl)-OH, H-H18(1-Me)-OH , H-His(1-Tos)-OH, 11-2.5-diiodo-1 Iis-(O11 and H-H18(3-Me)-OH.

Each parallel connecting unit or subunit thereof may be independently selected from the group consisting of glycine derivatives. Illustrative examples of glycine derivatives include, but are not limited to:

N-propargylglycine o

α-aminoglycine (protected or not), β-cyclopropylglycine, α-allylglycine and neopentylglycine.

- 32 046498

Each parallel linking unit or subunit thereof may be independently selected from the group consisting of glutamic acid and its derivatives. Illustrative examples of glutamic acid and its derivatives include, but are not limited to: glutamic acid (Glu), N-alkyl-glutamic acid, H-Glu(OtBu)-OH, H-','-hydroxy^1-OH, H- γ-methylene-Glu-OH, Η-γcarboxy-Glu(OtBu) ₂ -OH and pyroglutamic acid.

Each parallel linking unit or subunit thereof may be independently selected from the group consisting of glutamine and its derivatives. Illustrative examples of glutamine and its derivatives include, but are not limited to: glutamine (Gin), N-alkyl-glutamine, isoglutamine (HGlu-NH2), H-Gln(Trt)-OH and H-Gln(n'ionponnj)-OH .

Each parallel connecting unit or subunit thereof may be independently selected from the group consisting of phenylalanine (Phe) derivatives. Illustrative examples of phenylalanine derivatives include, but are not limited to: H-p-amino-Phe-OH, H-p-amino-Phe(Z)-OH, H-pbromo-Phe-OH, HH-p-carboxy-Phe( OtBu)-OH, H-p-carboxy-Phe-OH, H-p-cyano-Phe-OH, H-p-fluoro-PheOH, H-3,4-dichloro-Phe-OH, H-p-iodine -Phe-OH, N-p-nitro-Phe-OH, chloro-phenylalanine and β-homophenylalanine.

Each Parallel Linking Unit or subunit thereof may be independently selected from the group consisting of lysine and its derivatives. Illustrative examples of lysine and its derivatives include, but are not limited to: lysine (Lys), N-alkyl-lysine, H-Lys(Boc)-OH, H-Lys(Ac)-OH, H-LyB(Formyl)-OH , H-Lys(Me)2-OH, H-Lys(nicotinoyl)-OH, H-Lys(Me) ₃ -OH, H-trans-4,5-dehydro-LysOH, H-Lys(Alloc)-OH , Η-H-δ-hydroxy-Lys-OH, H-δ-hydroxy-Lys(Boc)-OH, H-Lys(acetamidoyl)-OH and H-Lys(isopropyl)-OH.

Each parallel connecting unit or subunit thereof may be independently selected from the group consisting of leucine derivatives. Illustrative examples of leucine derivatives include, but are not limited to, 4,5-dehydroleucine.

Each Parallel Connecting Unit or subunit thereof may be independently selected from the group consisting of methionine derivatives. Illustrative examples of methionine derivatives include, but are not limited to: methionine (Met), H-Met(=O)-OH and H-Met(=O) ₂ -OH, wherein the methionine side chain sulfur atom is in oxidized form.

Each Parallel Connecting Unit or subunit thereof may be independently selected from the group consisting of serine and its derivatives. Illustrative examples of serine and its derivatives include, but are not limited to: serine (Ser), N-alkyl-serine, H^r(Ac)-OH, H-Ser(tBu)-OH, H-Ser(Bzl)-OH , Η^^π-χλορ-Βζ1)-ΟΗ, H-β-(3,4-dihydroxyphenyl)-Ser-OH, H-β-(2-thienyl)-Ser-OH, isoserine N-alkyl-isoserine and 3-phenylisoserine.

Each Parallel Link Unit or subunit thereof may be independently selected from the group consisting of tyrosine and its derivatives. Illustrative examples of tyrosine and its derivatives include, but are not limited to: tyrosine (Tyr), N-alkyl-tyrosine, H-3, 5-dinitro-Tyr-OH, H-3-amino-Tyr-OH, H-3, 5-di6rom-Tyr-OH, H-3,5-diiodo-Tyr-OH, H-Tyr(Me)-OH, H-Tyr(tBu)-OH, HTyr(Boc)-OH, H-Tyr(Bzl )-OH, H-Tyr(Et)-OH, H-3-iodine-Tyr-OH and H-3-nitro-Tyr-OH.

Each parallel connecting unit or subunit thereof may be independently selected from the group consisting of threonine and its derivatives. Illustrative examples of threonine and its derivatives include, but are not limited to: threonine (Thr), N-alkyl-threonine, allo-threonine, H-Thr(Ac)OH, H-Thr(tBu)-OH and H-Thr(Bzl )-OH.

Each parallel connecting unit or subunit thereof may be independently selected from the group consisting of tryptophan and its derivatives. Illustrative examples of tryptophan and its derivatives include, but are not limited to: tryptophan (Tgr), N-alkyl-tryptophan, H-5-Me-TrpOH, H-5-hydroxy-Tgr-OH, H-4-Me-Tgr- OH, H-a-Me-Trp-OH, H-Trp(Boc)-OH, H-Trp(formyl)-OH and HTgr(mesitylene-2-sulfonyl)-OH.

Each parallel connecting unit or subunit thereof may be independently selected from the group consisting of proline and its derivatives. Illustrative examples of proline and its derivatives include, but are not limited to: proline (Pro), N-alkylproline, homoproline, thioproline, hydroxyproline (H-Hur-OH), H-Hur(1Bi)-OH. N-Nur(V/1)-0N. H-3,4-dehydro-Pro-OH, 4-ketoproline, αMe-Pro-OH and H-4-fluoro-Pro-OH.

Each parallel connecting unit or subunit thereof may be independently selected from the group consisting of ornithine and its derivatives. Illustrative examples of ornithine and its derivatives include, but are not limited to: ornithine (Ornithine), N-alkyl-ornithine, H-Orn(Boc)-OH, HOrn(Z)-OH, H-a-difluoro-Me-Orn- OH (eflornithine) and H-Orn(A11oc)-OH.

Each parallel connecting unit or subunit thereof may be independently selected from the group consisting of penicillamine and its derivatives. Illustrative examples of penicillamine and its derivatives include, but are not limited to: penicillamine, H-penicillamine (Ast)-OH (H-v,v-dimethylsue(Ast)-OH) and N-alkyl-penicillamine.

Each parallel connecting link or its subunit can be independently selected

- 33 046498 from the group consisting of β-alanine derivatives. Illustrative examples of β-alanine derivatives include, but are not limited to: dehydro-alanine.

Each parallel connecting unit or subunit thereof may be independently selected from the group consisting of aminoalkanoic acid derivatives. Illustrative examples of aminoalkanoic acid derivatives include, but are not limited to: 4(neopentyloxysulfonyl)aminobutyric acid, piperidylacetic acid, 3-aminopropionic acid and 3-amino-3-(3-pyridyl)propionic acid.

Each parallel connecting unit or subunit thereof may be independently selected from the group consisting of aminoalkynoic acid and its derivatives. Illustrative examples of aminoalkynoic acid and its derivatives include, but are not limited to, N-alkylaminoalkynic acid, 6-amino-4-hexynic acid, 6-(Boc-amino)-4-hexynic acid.

Each parallel connecting unit or subunit thereof may be independently selected from the group consisting of aminoalkanedioic acid and its derivatives. Illustrative examples of aminoalkanedioic acid and its derivatives include, but are not limited to, Nalkylaminoalkanedioic acid, 2-aminohexanedioic acid, 2-aminoheptanedioic acid, 2-aminooctanedioic acid (H-Asu-OH).

Each parallel connecting unit or subunit thereof may be independently selected from the group consisting of amino-heterocycloalkanoic acid and derivatives thereof. Illustrative examples of aminoheterocycloalkanoic acid and its derivatives include, but are not limited to: N-alkylaminoheterocycloalkanoic acids, 4-amino-1-methyl-1H-imidazole-2-carboxylic acid, 4-amino-1-methyl-1H-pyrrole-2-carboxylic acid , 4-aminopiperidine-4-carboxylic acid (H-Pip-OH; 1-protected or not), 3-amino-3-(3-pyridyl)propionic acid.

Each parallel connecting unit or subunit thereof may be independently selected from the group consisting of citrulline and its derivatives. Illustrative examples of citrulline and its derivatives include, but are not limited to: citrulline (cit), N-alkylcitrulline, thiocitrulline, S-methylthiocitrulline, and homocitrulline.

Illustrative examples of a statin and its derivatives include, but are not limited to: statin, N-alkyl-statin, cyclohexylstatin and phenylstatin.

Each parallel connecting unit or subunit thereof may be independently selected from the group consisting of diaminoalkanoic acid and its derivatives. Illustrative examples of diaminoalkanoic acid (Dab) and its derivatives include, but are not limited to: N-alkyldiaminoalkanoic acids, N-dialkylaminoalkanoic acids, α,γ-diaminobutyric acid (HDab-OH), H-Dab(Alloc)-OH, H-Dab(Boc)-OH, H-Dab(Z)-OH, α,β-diaminopropionic acid and its side chain protected variants.

An exemplary L ^P unit or subunit thereof, lysine or cysteine or penicillamine, is shown below. The wavy line indicates the sites of attachment to the PEG, the releasable complex group (directly or via a branching or drug attachment group), and the extending group (directly or via a branching or drug attachment group). L and D isomers of amino acids are suitable for use in the present invention.

- 34 046498

An example of a ligand-drug conjugate or drug-linker compound containing lysine as the LP unit is shown below, where Z, L, X, D, PEG, Z', p and PEG are as defined above. L and D isomers of amino acids are suitable for use in the present invention.

An example of a ligand-drug conjugate containing cysteine or penicillamine as the LP unit is shown below, where Z, L, X, D, Z', PEG and p are as defined above. The L and D isomers of amino acids are suitable for use in the present invention.

It will be appreciated that for some compounds of the present invention (e.g., linker intermediates and ligand-linker compounds), the parallel linking unit is capable of forming a covalent attachment to -X-D, but is not yet linked to -X-D, and the parallel

- 35 046498 This connecting unit will not yet be fully assembled into a ligand-drug conjugate and will thus include a functional group that is capable of interacting with a group present in the released complex group. An exemplary parallel connector containing a functional group for attachment is the following:

where the subscript n has a value from 1 to 4;

XP is selected from the group consisting of -O-, -NR-, -S-, -C (=O)- and -S(=O)-; And

R ¹ and R ² are independently selected from the group consisting of H, C _1-3 alkyl, phenyl or C ₂ -C5 heterocycle;

R ⁶ represents a protecting group, H, -C _1-3 alkyl or -OH, where the wavy lines indicate covalent attachment in the remainder of the linker intermediate or ligand-linker compound.

Particularly preferred reactive functional groups that provide XP are sulfhydryl groups to form disulfide bonds or thioether bonds. The functional group may be protected by a protecting group. LP may be a thiol-containing group (eg, a thiol-containing amino acid), and thus LP may be a protected thiol-containing amino acid, such as a protected cysteine, as shown below. Although the formula for the L-isomer of cysteine is given below, the D-isomer of cysteine is also suitable. In addition, the tert-butylthiol protecting group can be replaced by another suitable thiol protecting group. Thiol protecting groups include tert-butyl sulfide, nbutyl sulfide, n-propyl sulfide, methyl sulfide, phenyl sulfide, thiopyridyl, isopropyl sulfide, ethyl sulfide and cysteinyl.

LP may be a dipeptide comprising a protected thiol-containing amino acid, such as a protected cysteine-alanine dipeptide, as shown below:

where the wavy lines indicate the covalent attachment of LP in the rest of the linker intermediate

In preferred embodiments, the LP unit is selected to minimize or not contribute to increasing the hydrophobicity of the drug-linker portions of the ligand-drug conjugates.

In preferred aspects of the present invention, the LP group has a mass of no more than about 500 daltons, no more than about 200 daltons, from about 10 to about 500 daltons, or from about 10 to about 200 daltons.

At the ends of ligand-drug conjugates there are ligand groups, drug groups and PEG groups.

Ligand groups

In some embodiments of the invention, a ligand group is present. The ligand group (L-) is a targeting agent that specifically binds to the target molecule. The ligand may specifically bind to a cell component (cell-binding agent) or to other target molecules of interest. The ligand group acts to target and direct the delivery of the drug group to the con

- 36 046498 specific population of target cells with which the ligand group interacts. Ligands include, but are not limited to, proteins, polypeptides and peptides. Suitable ligand moieties include, for example, antibodies, e.g., full-length antibodies and antigen-binding fragments thereof, interferons, lymphokines, hormones, growth factors and colony-stimulating factors, vitamins, nutrient transport molecules (such as, but not limited to, transferrin), or any other cell-binding molecule or substance. The ligand may be, for example, a targeting protein other than the antibody. Alternatively, the ligand may be, for example, an antibody. Preferred ligands are high molecular weight proteins, for example, ligands having a molecular weight of at least about 80 kDa.

The ligand group may form a bond with an extending group. The ligand group must contain the required number of attachment sites for the linker drug, regardless of whether they are of natural or non-natural origin (eg, engineered). For example, for a subscript p value to be between 6 and 14, the ligand group must be capable of forming a bond with between 6 and 14 drug-linker groups. The attachment sites may be naturally occurring or engineered into the ligand. The ligand group may form a bond with the extending group of the linker group through a reactive or activated heteroatom or heteroatom-containing functional group of the ligand. Reactive or activated heteroatoms or heteroatom-containing functional group that may be present on the ligand group include sulfur (in one embodiment, from the sulfhydryl group of the ligand), C=O, or (in one embodiment, from the carbonyl, carboxyl or hydroxyl group of the ligand ) and nitrogen (in one embodiment, from the primary or secondary amino group of the ligand). Such heteroatoms may be present on the ligand in its natural state, for example, as a natural antibody, or may be introduced into the ligand through chemical modification or bioengineering methods.

In one embodiment, the ligand group contains a sulfhydryl group, and the ligand group is linked to the linker group through the sulfur atom of the sulfhydryl group.

In another embodiment, the ligand contains lysine residues that can react with activated esters (such esters include, but are not limited to, N-hydroxysuccinimide, pentafluorophenyl, and p-nitrophenyl esters) of the extending group of the linker and thereby form an amide bond , consisting of a nitrogen atom of the Ligand group and a C=O group of the linker.

In yet another aspect, the ligand group contains one or more lysine residues that can be chemically modified to introduce one or more sulfhydryl groups. The ligand group binds to the linker group through the sulfur atom of the sulfhydryl group. Reagents that can be used to modify lysines include, but are not limited to, Nsuccinimidyl S-acetylthioacetate (SATA) and 2-iminothiolane hydrochloride (Trout's Reagent).

In another embodiment, the ligand group may contain one or more carbohydrate groups, which may be chemically modified to contain one or more sulfhydryl groups. The ligand group binds to the extending group of the linker through the sulfur atom of the sulfhydryl group.

In yet another embodiment, the ligand group may contain one or more carbohydrate groups that can be oxidized to produce an aldehyde (-CHO) group (see, e.g., Laguzza, et al., 1989, J. Med. Chem. 32). 3):548-55).

The corresponding aldehyde can form a bond with the reactive site in the extending group.

Reactive sites on the extending group that can react with the carbonyl group on the ligand include, but are not limited to, hydrazine and hydroxylamine. Other protocols for modifying proteins to attach or associate drug groups are described in Coliganet al., Current Protocols in Protein Science, vol. 2, John Wiley & Sons (2002) (incorporated herein by reference).

The ligand group forms a bond with the reactive group in the extending group. A wide variety of reactive groups can be used and this will depend on the nature of the ligand group. The reactive group may be a maleimide that is present in the extending group (prior to attachment to L), and covalent attachment of L to the extending group is accomplished through the sulfhydryl group of the ligand to form a thio-substituted succinimide. The sulfhydryl group may be present on the ligand when the ligand is in a natural state, such as a naturally occurring moiety, or may be introduced into the ligand through chemical modification.

In yet another embodiment, the ligand is an antibody and the sulfhydryl group is formed by reduction of an interchain disulfide. Accordingly, in some embodiments, the linker group is conjugated to the cysteine moiety of the reduced interchain disulfides.

- 37 046498

In yet another embodiment, the ligand is an antibody, and the sulfhydryl group is chemically introduced into the antibody, for example by introducing a cysteine residue. Accordingly, in some embodiments, the extension group is conjugated to an embedded cysteine residue.

With regard to bioconjugates, it has been found that the site of drug conjugate can influence a number of parameters, including ease of conjugation, stability of the drug linker, effects on the biophysical properties of the resulting bioconjugates, and in-vitro cytotoxicity.

With respect to drug-linker stability, the site of conjugation of the drug-linker to the ligand may influence the ability of the conjugated drug-linker to undergo an elimination reaction and the transfer of the drug-linker from the ligand of the bioconjugate to an alternative reactive thiol present in the environment of the bioconjugate, such as, for example , a reactive thiol in albumin, free cysteine, or glutathione when found in plasma. Such regions include, for example, interchain disulfides, as well as certain cysteine engineered regions. The ligand-drug conjugates described herein can be conjugated to thiol residues at sites that are not sensitive to the elimination reaction (eg, position 239 according to the EU index based on the Kabat nomenclature), in addition to other sites.

When the conjugates include non-immunoreactive protein, polypeptide or peptide ligands in place of an antibody, useful non-immunoreactive protein, polypeptide or peptide ligands include, but are not limited to, transferrin, epidermal growth factors (EGF), bombesin, gastrin, gastrin-releasing peptide, platelet-derived growth factor, IL-2, IL-6, transforming growth factors (TGFs) such as TGF-α and TGF-β, vaccinia growth factor (VGF), insulin and insulin-like growth factors I and II, somatostatin, lectins and low-density lipoprotein apoprotein .

Particularly preferred ligands are antibodies, including intact antibodies. In fact, in any of the embodiments described herein, the ligand group may be an antibody. Beneficial polyclonal antibodies are heterogeneous populations of antibody molecules isolated from the serum of immunized animals. Beneficial monoclonal antibodies are homogeneous populations of antibodies to a specific antigenic determinant (eg, cancer cell antigen, viral antigen, microbial antigen, protein, peptide, carbohydrate, chemical, nucleic acid, or fragments thereof). A monoclonal antibody (mAb) to an antigen of interest can be produced using any method known in the art that produces antibody molecules using stable cell lines in culture.

Useful monoclonal antibodies include, but are not limited to, human monoclonal antibodies, humanized monoclonal antibodies, or chimeric human mouse (or other species) monoclonal antibodies. Antibodies include full-length antibodies and their antigen-binding fragments. Human monoclonal antibodies can be prepared by any of numerous methods known in the art (eg, Teng et al., 1983, Proc. Natl. Acad. Sci. USA. 80:7308-7312; Kozbor et al., 1983, Immunology Today 4:72-79; and Olsson et al., 1982, Meth. Enzymol. 92:3-16).

An antibody may be a functionally active fragment, derivative, or analogue of an antibody that immunospecifically binds to target cells (eg, cancer cell antigens, viral antigens, or microbial antigens) or other antibodies associated with tumor cells or matrix. In this regard, functionally active means that the fragment, derivative or analogue is capable of immunospecifically binding to target cells. To determine which CDR sequences bind to an antigen, synthetic peptides containing CDR sequences can be used in antigen binding assays using any binding assay method known in the art (eg, BIA nuclear assay) (see, e.g. , Kabatet al., 1991, Sequences of Proteins of Immunological Interest, Fifth Edition, National Institute of Health, Bethesda, Md; Rabat E et al., 1980, J. Immunology 125(3):961-969).

Other useful antibodies include antibody fragments such as, but not limited to, F(ab')2 fragments, Fab fragments, Fvs, single chain antibodies, diabodies, tribodies, tetrabodies, scFv, scFv-FV or any other molecule with the same specificity , like an antibody.

In addition, recombinant antibodies, such as chimeric and humanized monoclonal antibodies, including both human and non-human parts, which can be produced using standard recombinant DNA techniques, are useful antibodies. A chimeric antibody is a molecule in which different parts are derived from different animal species, such as, for example, containing a murine monoclonal-derived variable region and human immunoglobulin constant regions. (See, for example, US Pat. No. 4,816,567; and US Pat. No. 4,816,397, which are incorporated herein by reference in their entirety).

- 38 046498

Humanized antibodies are antibody molecules from a non-human species containing one or more complementary detectable regions (CDRs) from a non-human species and a framework region from a human immunoglobulin molecule. (See, for example, US Patent No. 5,585,089, which is incorporated herein by reference in its entirety.) Such chimeric and humanized monoclonal antibodies can be produced by recombinant DNA methods known in the art, for example, using the methods described in International Publication No. WO 87/02 671; European Patent Publication No. 0 184 187; European Patent Publication No. 0 171 496; European Patent Publication No. 0 173 494; International publication No. WO 86/01533; US Patent No. 4816567; European Patent Publication No. 012 023; Berter et al., 1988, Science 240:1041-1043; Liu et al., 1987, Proc. Natl. Acad. Sci. USA 84:3439-3443; Liu et al., 1987, J. Immunol. 139:3521-3526; Sun et al., 1987, Proc. Natl. Acad. Sci. USA 84:214-218; Nishimura et al., 1987, Cancer. Res. 47:999-1005; Wood et al., 1985, Nature 314:446-449; and Shaw et al., 1988, J. Natl. Cancer Inst. 80:1553-1559; Morrison, 1985, Science 229:1202-1207; Oi et al., 1986, BioTechniques 4:214; US Patent No. 5225539; Jones et al., 1986, Nature 321:552-525; Verhoeyan et al., 1988, Science 239:1534; and Beidler et al., 1988, J. Immunol. 141:4053-4060; each of which is incorporated herein by reference in its entirety.

Fully human antibodies are particularly desirable and can be produced using transgenic mice that are unable to express endogenous immunoglobulin heavy and light chain genes, but which can express human heavy and light chain genes.

Antibodies include analogs and derivatives that are also modified, i.e. by covalent attachment of any type of molecule, provided that such covalent attachment allows the antibody to retain its antigen-binding immunospecificity. For example, but not by way of limitation, antibody derivatives and analogues include those that have been further modified, e.g., by glycosylation, acetylation, PEGylation, phosphorylation, amidation, derivatization using known protecting/blocking groups, proteolytic cleavage, binding to the antibody cellular unit or another protein, etc. Any of numerous chemical modifications can be accomplished by known methods, including, but not limited to, specific chemical cleavage, acetylation, formylation, metabolic synthesis in the presence of tunicamycin, etc. In addition, the analogue or derivative may contain one or more non-naturally occurring amino acids.

Antibodies may have modifications (eg, substitutions, deletions, or additions) to amino acid residues that interact with Fc receptors. In particular, antibodies may contain modifications at amino acid residues identified as being involved in the interaction between the antiFc domain and the FcRn receptor (see, for example, International Publication No. WO 97/34631, which is incorporated herein by reference in its entirety).

Antibodies immunospecific for a cancer cell antigen can be produced commercially or can be produced by any method known to those skilled in the art, such as, for example, chemical synthesis or recombinant expression methods. The nucleotide sequence encoding antibodies immunospecific for a cancer cell antigen can be obtained, for example, from GenBank or a similar database, literature publications, or by routine cloning and sequencing.

In a specific embodiment, known antibodies can be used to treat cancer. Antibodies that are immunospecific for a cancer cell antigen can be produced commercially, or they can be produced by any method known to those skilled in the art, such as, for example, recombinant expression methods. The nucleotide sequence encoding antibodies immunospecific for a cancer cell antigen can be obtained, for example, from GenBank or a similar database, literature publications, or by routine cloning and sequencing.

In another specific embodiment, antibodies for treating an autoimmune disease are used in accordance with the compositions and methods of the present invention. Antibodies that are immunospecific for a cell antigen that is responsible for the production of autoimmune antibodies can be obtained from any organization (for example, from a university or company employee), or they can be obtained by any method known to those skilled in the art, such as, for example, chemical synthesis or recombinant expression methods.

In some embodiments, beneficial antibodies may bind to a receptor or receptor complex expressed on an activated lymphocyte. The receptor or receptor complex may include a member of the immunoglobulin gene superfamily, a member of the TNF receptor superfamily, an integrin, a cytokine receptor, a chemokine receptor, a major histocompatibility complex protein, a lectin, or a complement control protein.

In some aspects, the antibody will specifically bind to CD19, CD20, CD30, CD33, CD70, al(|)a-y-beta-6. Liv-1 or Lewis Y antigen.

- 39 046498

The Ahtu-CD30 antibody may be, for example, a chimeric AC10 antibody, brentuximab. The Ahtu-CD30 antibody may comprise a heavy chain variable region having the amino acid sequence set forth in SEQ ID NO: 1, a light chain variable region having the amino acid sequence set forth in SEQ ID NO: 2, a human gamma I constant region having the amino acid sequence set forth in SEQ ID NO: 7, and a human kappa constant region having the amino acid sequence set forth in SEQ ID NO: 8.

The Ahtu-CD30 antibody may be, for example, a humanized AC10 antibody. An anti-CD30 antibody may comprise a heavy chain variable region having the amino acid sequence set forth in SEQ ID NO: 9, a light chain variable region having the amino acid sequence set forth in SEQ ID NO: 10. The antibody may further comprise a human gamma I constant region, having the amino acid sequence set forth in SEQ ID NO: 7, optionally having a serine to cysteine substitution at position 239 (according to the EU index), and a human kappa constant region having the amino acid sequence set forth in SEQ ID NO: 8.

The Ahtu-CD70 antibody may be, for example, a humanized antibody (see, for example, US 2009/0148942). In an illustrative embodiment, the hhtu-CD70 antibody comprises a heavy chain variable region having the amino acid sequence set forth in SEQ ID NO: 3 and a light chain variable region having the amino acid sequence set forth in SEQ ID NO: 4.

The Ahtu-CD19 antibody may be, for example, a humanized antibody (see, for example, US 2009/0136526, incorporated herein by reference in its entirety and for all purposes). In an illustrative embodiment, the hBU12 antibody comprises a heavy chain variable region having the amino acid sequence set forth in SEQ ID NO: 5 and a light chain variable region having the amino acid sequence set forth in SEQ ID NO: 6.

The antibody may be a humanized hhtu-CD33 antibody (US 2013/0309223, incorporated herein by reference in its entirety and for all purposes), a humanized anti-Betab antibody (see, for example, WO 2013/123152, incorporated herein by reference in its entirety and for all purposes), a humanized anti-EA-1 antibody (see, for example, US 2013/0259860, incorporated herein by reference in its entirety and for all purposes), or a humanized AC10 antibody (see eg US 8257706, incorporated herein by reference in its entirety and for all purposes).

An example of attachment to a ligand is attachment via thioether bonds. Thioester linkages may include linkages through interchain disulfide bonds, embedded cysteine residues, and combinations thereof.

Drug groups

The effects of the present invention will be more pronounced in embodiments where the drugs are hydrophobic in nature. Accordingly, the drugs of the present invention are preferably hydrophobic in nature.

The drug group (D) may be a cytotoxic, cytostatic or immunosuppressive drug, also referred to in the present invention as a cytotoxic, cytostatic or immunosuppressive agent. The drug group contains an atom that can form a bond with the released complex group (X). In some embodiments, drug group D contains a nitrogen atom that can form a bond with a releasable complex group (X). In other embodiments, drug group D contains a carboxylic acid that can form a bond with a releasable complex group (X). In other embodiments, drug group D contains a sulfhydryl group that can form a bond with the releasable complex group X. In some other embodiments, drug group D contains a hydroxyl group or a ketone or alcohol that can form a bond with the releasable complex group X.

Useful classes of cytotoxic or immunosuppressive agents include, for example, antitubulin agents, DNA minor groove binding agents, DNA replication inhibitors, alkylating agents, antibiotics, antifolates, antimetabolites, chemotherapy sensitizing agents, topoisomerase inhibitors, vinca alkaloids, or the like. . Specific examples of useful classes of cytotoxic agents include, for example, DNA minor groove binding agents, DNA alkylating agents, and tubulin inhibitors. Examples of cytotoxic agents include, for example, auristatins, camptothecins, duocarmycins, etoposides, maytansines and maytansinoids, taxanes, benzodiazepines or benzodiazepine-containing drugs (eg, pyrrolo[1,4]-benzodiazepines (PBDs), indolinobenzodiazepines and oxazolidinobenzodiazepines) and vinca alkaloids. Selected benzodiazepine-containing medicinal products are described in

- 40 046498

WO 2010/091150, WO 2012/112708, WO 2007/085930 and WO 2011/023883.

In some embodiments, the cytotoxic agent is maytansine or a maytansinoid (eg, DM1, DM4) belonging to another group of anti-tubulin agents. (ImmunoGen, Inc.; see also Chari et al., 1992, Cancer Res. 52:127-131 and US Patent No. 8163888).

In some embodiments, the drug is a benzodiazepine (including benzodiazepine-containing drugs, e.g., pyrrolo[1,4]benzodiazepines (PBDs), indolinobenzodiazepines, and oxazolidinobenzodiazepines).

PBDs have a general structure:

but may differ in the number, type and position of substituents, both in their aromatic A rings and in the pyrrolo C rings, and in the degree of saturation of the C ring. The B-ring contains either an imine (N=C) or carbinolamine (NH-CH(OH)) or carbinolamine methyl ester (NH-CH(OMe)) at the N10-C11 position, which is the electrophilic center responsible for DNA alkylation. All known natural products have a ^-orientation at the chiral C11a position, which makes them bend to the right when looking from the C ring in the direction of the A ring. This gives them a suitable three-dimensional shape for isoshelicality of the minor groove of the B-form DNA, resulting in a precise fit at the binding site. The ability of PBDs to form a minor groove addition product allows them to interfere with DNA processing, hence making them possible to use as antitumor agents. The biological activity of these molecules can be enhanced, for example, by linking two PBD units together via their C8/C'-hydroxyl functionality via a flexible alkylene linker. PBD dimers are thought to form sequence-selective DNA lesions, such as the palindromic 5'-Pu-GATC-Py-3' interstrand cross-link, which is thought to be primarily responsible for their biological activity.

The drug group may be, for example, an auristatin or non-auristatin drug having a hydrophobicity comparable to or greater than monomethyl auristatin E. In some aspects, the drug is MMAE or auristatin having a hydrophobicity comparable to or greater than monomethyl auristatin E. Auristatin the drug may be covalently attached to the releasable complex group, for example through its N or C terminus. MMAE has a SlogP value of 2.59. In some aspects, drugs for use in the present invention will have a SlogP value of 1.5 or greater, 2.0 or greater, or 2.5 or greater. In some aspects, drugs for use in the present invention will have a SlogP value of (a) from about 1.5, about 2 or 2.5 to about 7, (b) from about 1.5, about 2 or 2.5 to about 6, (c) from about 1.5, about 2 or about 2.5 to about 5, (d) from about 1.5, about 2 or 2.5 to about 4, or (e) from about 1.5 , about 2 or about 2.5 to about 3.

The drug group may have the formula DE shown below, where the attachment to the releasable complex group is via the N terminus

where independently in each position

R ² is selected from the group consisting of H and C1-C8 alkyl;

R ³ is selected from the group consisting of H, C 1 -C 8 alkyl, C ₃ -C ₈ carbocycle, aryl, C 1 -C 8 alkylaryl, C 1 -C 8 alkyl (C ₃ -C ₈ carbocycle), C ₃ -C ₈ heterocycle and C1-C8 alkyl- (C ₃ -C ₈ heterocycle);

R ⁴ is selected from the group consisting of H, C 1 -C 8 alkyl, C ₃ -C ₈ carbocycle, aryl, C 1 -C 8 alkylaryl, C ₁ -C ₈ alkyl (C ₃ -C ₈ carbocycle), C ₃ -C ₈ heterocycle and C1-C8 alkyl- (C ₃ -C ₈ heterocycle);

R ⁵ is selected from the group consisting of H and methyl;

or R ⁴ and R ⁵ together form a carbocyclic ring and have the formula -(CR ^a R ^b )n-, where Ra and R ^b are independently selected from the group consisting of H, a C1-C8 alkyl and a C3- _C8 carbocycle, and n selected from the group consisting of 2, 3, 4, 5 and 6;

R ⁶ is selected from the group consisting of H and C1-C8 alkyl;

R ⁷ is selected from the group consisting of H, C1-C8 alkyl, _C3 - _C8 carbocycle, aryl, C1-C8 alkylaryl, C1-C8 alkyl-( _C3 - _C8 carbocycle), _C3 - _C8 heterocycle and C1-C8 alkyl-(C ₃ -C ₈ heterocycle);

each R ⁸ is independently selected from the group consisting of H, OH, C1- _C8 alkyl, _C3 - _C8 carbocycle and O-(C1- _C8 alkyl);

R ⁹ is selected from the group consisting of H and C1-C8 alkyl;

- 41 046498

R ¹⁸ is selected from the group consisting of -C(R ⁸ )2-C(R ⁸ )2-aputfa, -C(R ⁸ )2-C(R ⁸ )2-(C3-C8 heterocycle) and -C( R ⁸ )2-C (R ⁸ ) ₂ -(C ₃ -C ₈ carbocycle). MMAE conjugated through its N terminus is shown below

In some embodiments, the drug moiety is a vinca alkaloid, camptothecin, or anthracycline cytotoxic compound. Exemplary structures of such drug groups, when present in the X-D portion, are described herein for drug-linker intermediates.

There are many different assays that can be used to determine whether or not a ligand-drug conjugate has a cytostatic or cytotoxic effect on a cell line. In one example, a thymidine assay is used to determine whether or not a ligand-drug conjugate has a cytostatic or cytotoxic effect on a cell line. For example, cells at a density of 5000 cells/well are seeded in a 96-well plate and cultured for a 72-hour period and exposed to 0.5 μCi of ^3H -thymidine for the last 8 hours of that 72-hour period, and the inclusion of ^3H -thymidine in culture cells is measured in the presence and absence of a ligand-drug conjugate. A ligand-drug conjugate has a cytostatic or cytotoxic effect on a cell line if the cells in the culture have reduced ³ H-thymidine incorporation compared to cells of the same cell line cultured under the same conditions but not exposed to the ligand-drug conjugate.

In another example, to determine whether or not a ligand-drug conjugate has a cytostatic or cytotoxic effect on a cell line, cell viability is measured by determining the cell's uptake of a dye, such as neutral red, trypan blue, or ALAMAR™ blue (see, e.g., Page et al., 1993, Intl J of Oncology 3:473-476). In such an assay, cells are incubated in a medium containing a dye, the cells are washed, and the remaining dye, reflecting cellular uptake of the dye, is measured spectrophotometrically. The protein-bound dye sulforhodamine B (SRB) can also be used to measure cytotoxicity (Skehan et al., 1990, J. Nat'l Cancer Inst. 82:1107-12). Preferred ligand-drug conjugates include those that have an IC ₅₀ value (defined as the mAB concentration that produces 50% cell death) of less than 1000 ng/ml, preferably less than 500 ng/ml, more preferably less than 100 ng/ml and most preferably less than 50 or even less than 10 ng/ml per cell line.

General procedures for linking a drug to linkers are known in the art. See, for example, US Patent Nos. 8163888, 7659241, 7498298, US Publication No. US 20110256157 and International Application Nos. WO 2011023883 and WO 2005112919.

Polyethylene glycol group (PEG)

Polydisperse PEGs, monodisperse PEGs and discrete PEGs can be used to prepare the Compounds of the present invention. Polydisperse PEGs are a heterogeneous mixture of sizes and molecular weights, while monodisperse PEGs are typically purified from heterogeneous mixtures and therefore provide the desired individual chain length and molecular weight. Preferred PEG groups are discrete PEGs, compounds that are synthesized stepwise rather than through a polymerization process. Discrete PEGs provide a single molecule with a defined and defined chain length.

The PEG group provided in the present invention includes one or more polyethylene glycol chains. Polyethylene glycol chains can be linked together, for example, in a linear, branched or star configuration. Typically, at least one of the PEG chains is derivatized at one end for covalent attachment to the parallel connecting link. Examples of connections to a parallel connecting link include connections through non-conditionally cleavable bonds or through conditionally cleavable bonds. Examples of attachments include attachment through an amide bond, ether bonds, ester bonds, hydrazone bonds, oxime bonds, disulfide bonds, peptide bonds or triazole bonds. In some aspects, the attachment to L ^P is accomplished through a non-conditionally cleavable bond. In some aspects, the attachment to L ^P is not via an ester bond, a hydrazone bond, an oxime bond, or a disulfide bond. In some aspects, the attachment to L ^P is not via a hydrazone bond.

A conditionally cleavable bond refers to a bond that is not inherently sensitive to cleavage when circulating in plasma, but is sensitive to cleavage in the intracellular or intratumoral environment. A non-conditionally cleavable bond is one that is not inherently susceptible to cleavage in any biological environment. Chemical

- 42 046498 hydrazone hydrolysis, disulfide reduction, and enzymatic cleavage of a peptide bond or glycosidic bond are examples of conditionally cleavable bonds.

The PEG group is directly attached to the ligand-drug conjugate (or its intermediate) on a parallel link. The other end (or ends) of the PEG group will be free and unrestricted and may take the form of a methoxy, carboxylic acid, alcohol or other suitable functional group. A methoxy, carboxylic acid, alcohol, or other suitable functional group acts as a cap for the terminal PEG subunit of the PEG Group. By non-restricted it is meant that the PEG group will not be attached at this non-restricted site to a drug group, to a ligand group, or to a linking moiety that binds a drug group and/or a ligand group. For those embodiments where the PEG group includes more than one PEG chain, the multiple PEG chains may be the same or different chemical groups (eg, PEGs having different molecular weights or numbers of subunits). These multiple PEG chains are attached to a parallel connecting link at one attachment site. Those skilled in the art will appreciate that the PEG moiety, in addition to comprising polyethylene glycol repeat subunits, may also contain non-PEG material (eg, to assist in linking multiple PEG chains to each other or to assist in linking to a parallel link). Non-PEG material refers to the atoms in the PEG group that are not part of the repeating -CH2CH2O- subunits. In the embodiments presented herein, the PEG group may include two monomeric PEG chains linked to each other through non-PEG elements. In other embodiments provided herein, the PEG moiety may include two linear PEG chains attached to a central core that is attached to a parallel connecting link (ie, the PEG moiety is itself branched).

There are many methods for attaching PEG known to those skilled in the art [see, for example, Goodson, et al. (1990) Bio/Technology 8:343 (PEGylation of interleukin-2 at its glycosylation site after site-directed mutagenesis); EP 0401384 (PEG binding to G-CSF); Malik, et al., (1992) Exp. Hematol. 20:1028-1035 (PEGylation of GM-CSF using tresyl chloride); ACT Publication No. WO 90/12874 (PEGylation of erythropoietin containing a recombinantly inserted cysteine residue using a cysteine-specific mPEG derivative); US patent No. 5757078 (PEGylation of EPO peptides); US Patent No. 5,672,662 (poly(ethylene glycol) and related polymers monosubstituted with propionic or butanoic acid and their functional derivatives for biotechnical applications); US Patent No. 6,077,939 (PEGylation of the N-terminal alpha carbon of a peptide); Veronese et al., (1985) Appl. Biochem. Bioechnol 11:141-142 (PEGylation of the N-terminal α-carbon of a peptide using PEG-nitrophenyl carbonate (PEG-NPC) or PEGtrichlorophenyl carbonate); and Veronese (2001) Biomaterials 22:405–417 (review article on PEGylation of peptides and proteins)].

For example, PEG can be covalently linked to amino acid residues through a reactive group. Reactive groups are those to which the activated PEG molecule can bind (eg, a free amino or carboxyl group). For example, N-terminal amino acid residues and lysine (K) residues contain a free amino group; and The terminal amino acid residues contain a free carboxyl group. Sulfhydryl groups (eg, those present on cysteine residues) can also be used as a reactive group for PEG attachment. In addition, enzymatic methods have been described to introduce activated groups (eg, hydrazide, aldehyde and aromatic amino groups) specifically at the C-terminus of the polypeptide (see Schwarz, et al. (1990) Methods Enzymol. 184:160; Rose, et al. (1991) Bioconjugate Chem. 2:154; and Gaertner, et al. (1994) J. Biol. Chem. 269:7224].

In some embodiments, PEG molecules can be attached to amino groups using methoxylated PEG (mPEG) containing various reactive groups. Non-limiting examples of such reactive groups include succinimidyl succinate (SS), succinimidyl carbonate (SC), mPEG imidate, para-nitrophenyl carbonate (NPC), succinimidyl propionate (SPA) and cyanuric acid chloride. Non-limiting examples of such mPEGs include mPEG-succinimidyl succinate (mPEG-SS), mPEG ₂ -succinimidyl succinate (mPEG ₂ -SS); mPEG succinimidyl carbonate (mPEG-SC), mPEG ₂ -succinimidyl carbonate (mPEG ₂ ^C); mPEG-imidate, mPEG-para-nitrophenyl carbonate (mPEG-NPC), mPEG-imidate; mPEG ₂ -para-nitrophenyl carbonate (mPEG ₂ -NiC); mPEG-succinimidylpropionate (mPEG-SPA); mPEG ₂ -succinimidyl propionate (mPEG, --SPA); mPEG-Nhydroxysuccinimide (mPEG-NHS); mPEG ₂ -N-hydroxysuccinimide (mPEG ₂ -HNi); mPEG-cyanuric acid chloride; mPEG ₂ -cyanuric acid chloride; mPEG ₂ -lysinol-NaC and mPEG ₂ -LysNHS.

Typically, at least one of the PEG chains that make up the PEG group is functionalized so that it can be attached to a parallel connecting link. Functionalization can be accomplished, for example, through an amine, thiol, NHS ester, maleimide, alkyne, azide, carbonyl, or other functional group. The PEG group may further include

- 43 046498 non-PEG material (ie material not consisting of -CH2CH2O-) as facilitating the binding to a parallel connecting link or as promoting the binding of two or more PEG chains.

A wide variety of polyethylene glycol (PEG) compounds can be used, and essentially any suitable PEG reactive reagent can be used. In some embodiments, the reactive PEG reagent will result in the formation of a carbamate or amide bond when attached to the LP. The following PEG reagents are useful in various embodiments: m1 GE1' ₂ -NI IS, mPE^-ALD, branched PEG, mPEG(MAB)2, mPEG ₂ (MAL), mPEG^H ₂ , mPEG-SPA, mPEG -SBA, mPEG-thioesters, mPEG-two ester groups, mPEG-BTS, mPEG-ButyrALD, mPEG-ASET, heterofunctional PEGs (KH ₂ -PEGCOOH, Boc-PEG-NHS, Fmoc-PEG-NHS, NHS-PEG-VS , NHS-PEG-MAL), PEG acrylates (ACRL-PEGNHS), PEG phospholipids (e.g. mPEG-DSPE), branched PEGs of the SUNBRITE™ series, including the GL series of glycerol-based PEGs, activated by chemical methods selected by specialists in given area, any of the SUNBRITE activated PEGs (including but not limited to carboxyl-PEG groups, p-NP-PEG groups, tresyl-PEG groups, aldehyde PEG groups, acetal-PEG groups, amino-PEG groups, thiol-PEG groups , maleimido-PEG groups, hydroxyl-PEG-amine, amino-PEG-COOC hydroxyl-PEG-aldehyde, carboxylic acid-PEG anhydride, functionalized PEG-phospholipid and other similar and/or suitable reactive PEGs selected by those skilled in the art for specific purposes and applications.

The addition of a PEG moiety may have two potential effects on the pharmacokinetics of the resulting ligand-drug conjugate. The desired effect is to reduce clearance (and resultantly increase exposure), which occurs as a result of reducing non-specific interactions induced by the exposed hydrophobic elements of the drug linker. The second effect is an undesirable effect and is a decrease in volume and rate of distribution, which may occur as a result of an increase in the molecular weight of the ligand-drug conjugate. Increasing the number of PEG subunits increases the hydrodynamic radius of the conjugate, resulting in reduced diffusive transfer ability. In turn, reduced diffusive transport capacity may reduce the ability of the ligand-drug conjugate to penetrate the tumor (Schmidt and Wittrup, Mol Cancer Ther 2009;8:28612871). Because of these two competing pharmacokinetic effects, it is desirable to use a PEG that is large enough to reduce LDC clearance, thereby increasing plasma levels of the drug, but not large enough to greatly reduce its ability to diffusively transport, which could reduce the ability of the ligand-conjugate. the drug to reach the desired target cell population. For guidelines on selecting the optimal PEG size for a particular linker drug, see the examples (eg, Examples 1, 18, and 21).

In one group of embodiments, the PEG group includes at least 6 subunits, at least 7 subunits, at least 8 subunits, at least 9 subunits, at least 10 subunits, at least 11 subunits, at least 12 subunits , at least 13 subunits, at least 14 subunits, at least 15 subunits, at least 16 subunits, at least 17 subunits, at least 18 subunits, at least 19 subunits, at least 20 subunits, by at least 21 subunits, at least 22 subunits, at least 23 subunits, or at least 24 subunits. In the context of the present invention, a subunit when referring to a PEG group refers to a polyethylene glycol subunit having the formula -£-(CH ₂ CH ₂ O) -

In some such embodiments, the PEG moiety includes no more than about 72 subunits.

In one set of embodiments, the PEG moiety includes one or more linear PEG chains, each of which contains at least 2 subunits, at least 3 subunits, at least 4 subunits, at least 5 subunits, at least 6 subunits, at least 7 subunits, at least 8 subunits, at least 9 subunits, at least 10 subunits, at least 11 subunits, at least 12 subunits, at least 13 subunits, at least 14 subunits, at least 15 subunits, at least 16 subunits, at least 17 subunits, at least 18 subunits, at least 19 subunits, at least 20 subunits, at least 21 subunits, at least 22 subunits, at least 23 subunits or at least 24 subunits. In preferred embodiments, the PEG group includes a total of at least 6 subunits, at least 8, at least 10 subunits, or at least 12 subunits. In some such embodiments, the PEG group includes a total of no more than about 72 subunits, preferably a total of no more than about 36 subunits.

In another group of embodiments, the PEG group includes in general from 4 to 72, from 4 to 60, from

- 44 046498 to 48, from 4 to 36 or from 4 to 24 subunits, from 5 to 72, from 5 to 60, from 5 to 48, from 5 to 36 or from 5 to 24 subunits, from 6 to 72, from 6 up to 60, from 6 to 48, from 6 to 36 or from 6 to 24 subunits, from 7 to 72, from 7 to 60, from 7 to 48, from 7 to 36 or from 7 to 24 subunits, from 8 to 72, from 8 to 60, from 8 to 48, from 8 to 36 or from 8 to 24 subunits, from 9 to 72, from 9 to 60, from 9 to 48, from 9 to 36 or from 9 to 24 subunits, from 10 to 72, from 10 to 60, from 10 to 48, from 10 to 36 or from 10 to 24 subunits, from 11 to 72, from 11 to 60, from 11 to 48, from 11 to 36 or from 11 to 24 subunits, from 12 to 72, 12 to 60, 12 to 48, 12 to 36 or 12 to 24 subunits, 13 to 72, 13 to 60, 13 to 48, 13 to 36 or 13 to 24 subunits , from 14 to 72, from 14 to 60, from 14 to 48, from 14 to 36 or from 14 to 24 subunits, from 15 to 72, from 15 to 60, from 15 to 48, from 15 to 36 or from 15 to 24 subunits, 16 to 72, 16 to 60, 16 to 48, 16 to 36 or 16 to 24 subunits, 17 to 72, 17 to 60, 17 to 48, 17 to 36 or 17 to 24 subunits, 18 to 72, 18 to 60, 18 to 48, 18 to 36 or 18 to 24 subunits, 19 to 72, 19 to 60, 19 to 48, 19 to 36 or from 19 to 24 subunits, from 20 to 72, from 20 to 60, from 20 to 48, from 20 to 36 or from 20 to 24 subunits, from 21 to 72, from 21 to 60, from 21 to 48, from 21 up to 36 or from 21 to 24 subunits, from 22 to 72, from 22 to 60, from 22 to 48, from 22 to 36 or from 22 to 24 subunits, from 23 to 72, from 23 to 60, from 23 to 48, from 23 to 36 or from 23 to 24 subunits, or from 24 to 72, from 24 to 60, from 24 to 48, from 24 to 36 or from 24 subunits.

In another group of embodiments, the PEG moiety comprises one or more linear PEG chains containing a total of 4 to 72, 4 to 60, 4 to 48, 4 to 36, or 4 to 24 subunits, 5 to 72, 5 to 60, 5 to 48, 5 to 36 or 5 to 24 subunits, 6 to 72, 6 to 60, 6 to 48, 6 to 36 or 6 to 24 subunits, 7 to 72 , from 7 to 60, from 7 to 48, from 7 to 36 or from 7 to 24 subunits, from 8 to 72, from 8 to 60, from 8 to 48, from 8 to 36 or from 8 to 24 subunits, from 9 to 72, 9 to 60, 9 to 48, 9 to 36 or from 9 to 24 subunits, from 10 to 72, 10 to 60, 10 to 48, 10 to 36 or from 10 to 24 subunits, from 11 to 72, 11 up to 60, 11 to 48, 11 to 36 or from 11 to 24 subunits, from 12 to 72, from 12 to 60, from 12 to 48, from 12 to 36 or from 12 to 24 subunits, from 13 to 72, 13 to 60, from 13 to 48, from 13 to 36 or from 13 to 24 subunits, from 14 to 72, from 14 to 60, from 14 to 48, from 14 to 36 or from 14 to 24 subunits, from 15 to 72, from 15 to 60, 15 to 48, 15 to 36 or 15 to 24 subunits, 16 to 72, 16 to 60, 16 to 48, 16 to 36 or 16 to 24 subunits, 17 to 72 , from 17 to 60, from 17 to 48, from 17 to 36 or from 17 to 24 subunits, from 18 to 72, from 18 to 60, from 18 to 48, from 18 to 36 or from 18 to 24 subunits, from 19 up to 72, from 19 to 60, from 19 to 48, from 19 to 36 or from 19 to 24 subunits, from 20 to 72, from 20 to 60, from 20 to 48, from 20 to 36 or from 20 to 24 subunits, 21 to 72, 21 to 60, 21 to 48, 21 to 36 or 21 to 24 subunits, 22 to 72, 22 to 60, 22 to 48, 22 to 36 or 22 to 24 subunits, from 23 to 72, from 23 to 60, from 23 to 48, from 23 to 36 or from 23 to 24 subunits, or from 24 to 72, from 24 to 60, from 24 to 48, from 24 to 36 or from 24 subunits.

In another group of embodiments, the PEG moiety is a derivatized linear single PEG chain containing at least 2 subunits, at least 3 subunits, at least 4 subunits, at least 5 subunits, at least 6 subunits, at least 7 subunits, at least 8 subunits, at least 9 subunits, at least 10 subunits, at least 11 subunits, at least 12 subunits, at least 13 subunits, at least 14 subunits, at least 15 subunits , at least 16 subunits, at least 17 subunits, at least 18 subunits, at least 19 subunits, at least 20 subunits, at least 21 subunits, at least 22 subunits, at least 23 subunits, or at least 24 subunits.

In another group of embodiments, the PEG moiety is a derivatized linear single PEG chain containing 6 to 72, 6 to 60, 6 to 48, 6 to 36, or 6 to 24 subunits, 7 to 72, 7 to 60, from 7 to 48, from 7 to 36 or from 7 to 24 subunits, from 8 to 72, from 8 to 60, from 8 to 48, from 8 to 36 or from 8 to 24 subunits, from 9 to 72, 9 up to 60, 9 to 48, 9 to 36 or from 9 to 24 subunits, from 10 to 72, 10 to 60, 10 to 48, 10 to 36 or from 10 to 24 subunits, from 11 to 72, from 11 to 60, from 11 to 48, from 11 to 36 or from 11 to 24 subunits, from 12 to 72, from 12 to 60, from 12 to 48, from 12 to 36 or from 12 to 24 subunits, from 13 to 72, from 13 to 60, from 13 to 48, from 13 to 36 or from 13 to 24 subunits, from 14 to 72, from 14 to 60, from 14 to 48, from 14 to 36 or from 14 to 24 subunits, from 15 to 72, from 15 to 60, 15 to 48, 15 to 36 or 15 to 24 subunits, 16 to 72, 16 to 60, 16 to 48, 16 to 36 or 16 to 24 subunits, 17 to 72 , from 17 to 60, from 17 to 48, from 17 to 36 or from 17 to 24 subunits, from 18 to 72, from 18 to 60, from 18 to 48, from 18 to 36 or from 18 to 24 subunits, from 19 up to 72, from 19 to 60, from 19 to 48, from 19 to 36 or from 19 to 24 subunits, from 20 to 72, from 20 to 60, from 20 to 48, from 20 to 36 or from 20 to 24 subunits, 21 to 72, 21 to 60, 21 to 48, 21 to 36 or 21 to 24 subunits, 22 to 72, 22 to 60, 22 to 48, 22 to 36 or 22 to 24 subunits, from 23 to 72, from 23 to 60, from 23 to 48, from 23 to 36 or from 23 to 24 subunits, or from 24 to 72, from 24 to 60, from 24 to 48, from 24 to 36 or from 24 subunits.

In another group of embodiments, the PEG moiety is a derivatized linear single PEG chain containing 2 to 72, 2 to 60, 2 to 48, 2 to 36, or 2 to 24 subunits, 2 to 72, 2 to 60, from 2 to 48, from 2 to 36 or from 2 to 24 subunits, from 3 to 72, from 3 to 60, from 3 to 48, from 3 to 36 or 3 to 24 subunits, from 3 to 72, from 3 up to 60, from 3 to 48, from 3 to 36 or from 3 to 24 subunits, from 4 to 72, from 4 to 60, from 4 to 48, from 4 to 36 or from 4 to 24 subunits, from 5 to 72, from 5 to 60,

- 45 046498 from 5 to 48, from 5 to 36 or from 5 to 2 4 subunits.

Examples of linear PEG groups that can be used in any of the embodiments presented in this application are the following:

-£>-R ²⁰ (CH ₂ CH ₂ O) _P R ²¹ -R ²⁰ (CH2CH2O) P. R ²² -(CH2CH ₂ O) _P. R ²¹ -R ²⁰ -(CH ₂ CH ₂ O) _P .

where the wavy line indicates the area of connection to the parallel connecting link,

R ²⁰ represents a PEG addition group,

R ²¹ represents a PEG blocking group;

R ²² is a PEG-binding group (ie, for linking multiple chains of PEG subunits together) n is independently selected from 2 to 72 (preferably 4 to 72, more preferably 6 to 72, 8 to 72, 10 to 72, 12 to 72 or 6 to 24);

e has a value from 2 to 5, each n' is independently selected from values from 1 to 72. In preferred embodiments, there are at least 6, preferably at least 8, at least 10, or at least 12 PEG subunits present in the PEG group . In some embodiments, no more than 72 or 36 PEG subunits are present in the PEG group.

In preferred embodiments, n is 8 or about 8, 12 or about 12, 24 or about 24.

The PEG attachment group is part of the PEG group and acts as a linking PEG group with a parallel connecting link. In this regard, the parallel connecting link contains a functional group that forms a bond with the PEG group. Functional groups for attaching a PEG group to a parallel connecting unit include sulfhydryl groups to form disulfide bonds or thioether bonds, aldehyde, ketone or hydrazine groups to form hydrazone bonds, hydroxylamine to form oxime bonds, carboxyl or amino groups to form peptide bonds, carboxyl or hydroxy groups to form ester bonds, sulfonic acids to form sulfonamide bonds, alcohols to form carbamate bonds, and amines to form sulfonamide bonds or carbamate bonds or amide bonds. Accordingly, the PEG group may be attached to the Parallel Linking Unit, for example, through a disulfide, thioether, hydrazone, oxime, peptide, ester, sulfonamide, carbamate or amide linkage. Typically, a PEG Addition Group is the product of a cycloaddition, addition, addition/elimination, or displacement reaction that occurs when a PEG group is added to a Parallel Connector.

The PEG Linking Group is part of the PEG group and is a non-PEG element that functions by linking two or more chains of repeating CH ₂ CH ₂ O subunits. In illustrative embodiments, the PEG-linking Group R ²² is -C _1-10 alkyl-C^yKI-, -C _1-10 alkyl-NH-C(O)-, -C _2-10 alkyl-NH-, -C _2-10 alkyl-O-, -C _1-10 alkyl-S- or -C _2-10 alkylNH-.

In illustrative embodiments, the PEG attachment group R ²⁰ is -C(O)-, -O-, -S-, -S(O)-, -NH-, -C(O)O-, -C(O) C1-yualkyl, -C(O)C1.yualkyl-O-, -C(O)C1-10alkyl-CO2-, -C(O)C1yualkyl-NH-, -C(O)C1-yualkyl^-, - C(O)C1.ualkyl-C(O) -NH-, -C(O)C1.ualkyl-KH-C(O)-, -C1- _Iu alkyl, -C _1-10 alkyl-O-, - C _1-10 alkyl-CO ₂ -, -C _1-10 alkyl-KH-, -C _1-10 alkyl^-, -C _1-10 alkyl-C(O)-KH-, -C _1-10 alkyl -KH-C(O)-, -CH ₂ CH^O ₂ -C _1-10 alkyl-, -CH ₂ C(O)-C _1-10 alkyl-, =N-(O or K)-C _{1 -10} alkyl-O-, =N-(O or K)-C _1-10 alkyl-KH-, =N(O or Yu-Cyoalkyl-COg-, =N-(O or N) -C _1-10 alkyl^O

each R ²¹ independently represents -C _1-10 alkyl, -C _2-10 alkyl-CO ₂ H, -C _2-10 alkyl-OH, -C _2-10 alkyl-MIg, C _2-10 alkyl-H2H^ alkyl) or C _2-10 alkyl-^C^ alkyl) ₂ ; and each R ²² independently represents -C _1-10 alkyl-C(O)-NH-, -C _1-10 alkyl-NH-C(O)-, -C _2-10 alkyl-NH-, -C _{2 -10} alkyl-O-, -C _1-10 alkyl-S- or -C _2-10 alkyl-NH-.

In some embodiments, R ²⁰ is -NH-, -C(=O)-, triazole-linked groups, or -S- or maleimido-linked groups, such as

- 46 046498

where the wavy line indicates the site of attachment to the parallel connecting link, and the asterisk indicates the site of connection in the PEG group. In some such aspects, R ²¹ is C _1-10 alkyl, -C _2-10 alkyl-CO ₂ H, -C _2-10 alkyl-OH, or -C _2-10 alkyl-MH ₂ .

Exemplary linear PEG groups that can be used in any of the embodiments presented herein are the following:

NH-(CH ₂ CH2O) _n -CH2CH2CO ₂ H

-^-NH-(CH ₂ CH ₂ O) _n -CH ₂ CH ₂ C(=O)NH-(CH ₂ CH ₂ O)-CH ₂ CH ₂ CO ₂ H o

<> ¹¹

-§-C—(CH ₂ CH ₂ O) _p -CH3

-^-NH-(CH ₂ CH ₂ O) _n -CH ₂ CH ₂ NH -(CH ₂ CH ₂ O)-CH ₂ CH ₂ CO ₂ H where the wavy line indicates the site of attachment to the Parallel Connecting Link, and each n is independently selected from the values 4 to 72, 6 to 72, 8 to 72, 10 to 72, 12 to 72, 6 to 24, or 8 to 24. In some aspects, n has a value of about 8, about 12 or about 24.

As described herein, the PEG moiety is selected such that it improves the clearance of the resulting ligand-drug conjugate but does not significantly affect the ability of the conjugate to penetrate the tumor. In embodiments wherein the drug group and the released complex group of the ligand-drug conjugate have hydrophobicity. comparable to the hydrophobicity of the maleimido-glucuronide MMAE drug linker (as shown in the examples), the PEG moiety selected for use will preferably contain from 8 subunits to about 24 subunits, more preferably about 12 subunits. In embodiments where the drug moiety and the released complex moiety of the conjugate have a hydrophobicity greater than that of the maleimido glucuronide MMAE drug linker, a PEG moiety with more subunits may be selected. The methods shown in the Examples section can be used to determine the ideal number of subunits for a particular linker drug.

In preferred embodiments of the present invention, the PEG group has a mass of from about 300 to about 5 kDa; from about 300 Da to about 4 kDA; from about 300 Da to about 3 kDa; from about 300 Da to about 2 kDa; or from about 300 Da to about 1 kDa. In some such aspects, the PEG group contains at least 6 subunits or at least 8, 10 or 12 subunits. In some such aspects, the PEG group contains at least 6 subunits or at least 8, 10 or 12 subunits, but no more than 72 subunits, preferably no more than 36 subunits.

In preferred embodiments of the present invention, no other PEG subunit other than the PEG group is present in the linker drug (ie, no PEG subunits are present in any other components of the conjugates and linkers provided herein). In other aspects of the present invention, in addition to PEG groups, the drug linker contains no more than 8, no more than 7, no more than 6, no more than 5, no more than 4, no more than 3, no more than 2, or no more than 1 other polyethylene glycol subunits (i.e., no more than 8, 7, 6, 5, 4, 3, 2 or 1 other polyethylene glycol subunits in the other components of the conjugates and linkers provided herein). The components include an extension group, a parallel connecting unit, a drug group, a branching group and a release complex group.

It should be understood that when referring to a PEG subunit, and depending on the context, the number of subunits may be an average number, for example, when referring to a population of ligand-drug conjugates or intermediates and when using polydisperse PEGs.

Expansion group:

The Extension Group (-Z-) performs the function of linking the ligand group to the parallel connecting link. In this regard, the extending group contains a functional group that can form a bond with a functional group of the ligand. The extension group also contains a functional group that can form a connection with the functional group of either an optional branching group or a parallel connecting link. In a ligand-drug conjugate and intermediates containing more than one drug group

- 47 046498 of a potent agent on a PEG group, the extending group will contain a functional group that can form a bond with a functional group of the ligand, and a functional group that can form a bond with a branching group, a parallel linking unit or a drug attachment group. Useful functional groups that may be present on the ligand group, either naturally or through chemical manipulation, include, but are not limited to, sulfhydryl (-SH), amino, hydroxyl, carboxy, anomeric carbohydrate hydroxyl and carboxyl. In one aspect, the functional groups on the ligand are sulfhydryl and amino. an extending group may include, for example, a maleimide group, an aldehyde, a ketone, a carbonyl, or a haloacetamide for attachment to a ligand group.

In some aspects, the extension group of the drug-linker compound or linker intermediate contains an electrophilic group that is capable of reacting with a nucleophilic group present on the ligand (eg, an antibody). Useful nucleophilic groups on the ligand include, but are not limited to, sulfhydryl, hydroxyl and amino groups. The heteroatom of the nucleophilic group of the ligand is capable of reacting with the electrophilic group on the extending group and forms a covalent bond with the extending group. Useful electrophilic groups include, but are not limited to, maleimide and haloacetamide groups. For an antibody as a ligand, the electrophilic group provides a convenient attachment site for the antibody for those antibodies that contain an accessible nucleophilic group.

In another embodiment, the extending group contains a reactive site that contains a nucleophilic group that is capable of reacting with an electrophilic group present on the ligand (eg, an antibody). Useful electrophilic groups on the ligand include, but are not limited to, aldehyde and ketone and carbonyl groups. The heteroatom of the nucleophilic group of the extending group can interact with the electrophilic group of the ligand and form a covalent bond with the antibody. Useful nucleophilic groups on the extending group include, but are not limited to, hydrazide, hydroxylamine, amino, hydrazine, thiosemicarbazone, hydrazine carboxylate and aryl hydrazide. For an antibody as a ligand, the electrophilic group of the antibody provides a convenient site for attachment to a nucleophilic extending group.

In some aspects, conjugates can be prepared by using a portion of an extending group containing a reactive site for binding to a parallel linking unit and introducing another portion of an extending group containing a reactive site for a ligand group. In one aspect, the extending group contains a reactive region that contains an electrophilic group that is capable of reacting with a nucleophilic group present on a ligand, such as an antibody. The electrophilic group provides a convenient site for attachment of a ligand (eg, antibody). Useful nucleophilic groups in an antibody include, but are not limited to, sulfhydryl, hydroxyl and amino groups. The heteroatom of the nucleophilic group of an antibody is capable of reacting with the electrophilic group of the extending group and forms a covalent bond with the extending group. Useful electrophilic groups include, but are not limited to, maleimide and haloacetamide groups and NHS esters.

In another embodiment, the extending group contains a reactive site that contains a nucleophilic group that is capable of reacting with an electrophilic group present on the ligand. An electrophilic group on a ligand (eg an antibody) provides a convenient site for attachment to an extender group. Useful electrophilic groups on an antibody include, but are not limited to, aldehyde and ketone carbonyl groups. The heteroatom of the nucleophilic group of the extending group can interact with the electrophilic group on the antibody and form a covalent bond with the antibody. Useful nucleophilic groups in the extender group include, but are not limited to, hydrazide, oxime, amino, hydrazine, thiosemicarbazone, hydrazine carboxylate and arylhydrazide.

In some embodiments, the extending group forms a bond to the sulfur atom of the ligand through the maleimide group of the extending group. The sulfur atom may come from the sulfhydryl group of the ligand. Representative extension groups of this embodiment include the bracketed groups of Formulas XVa and XVb, where the wavy line indicates attachment in a ligand-drug conjugate or intermediates thereof, and R ¹⁷ is -C ₁ -C _1o alkylene, C ₁ -C _1o heteroalkylene-, -C ₃ -C ₈ carbocyclo-, -O-(C ₁ -C ₈ alkyl)-, -arylene-, -C ₁ -C _1o alkylene-arylene-, -arylene-C1-C ₁₀ alkylene-, -C1-C ₁₀ alkylene-(C ₃ -C ₈ carbocyclo)-, -(C ₃ -C8 carbocyclo)-C1-C ₁₀ alkylene-, -C3-C8 heterocyclo-, -C1-C ₁₀ alkylene-(C ₃ —C ₈ heterocyclo)-, -(C3-C8 heterocyclo)-C1-C ₁₀ alkylene-, -C ₁ -C ₁₀ alkylene-C(=O)-, C1-C ₁₀ heteroalkylene-C(=O)-, -C3—C8 carbocyclo-C(=O)-, -O-(C1-C ₈ alkyl)C(=O)-, -arylene-C(=O)-, -C1-C ₁₀ alkylene-arylene-C (=O)-, -arylene-C1-C ₁₀ alkylene-C(=O)-, -C1-C ₁₀ alkylene(C3-C8 carbocyclo)-C(=O)-, - (C3-C8 carbocyclo)- C1-C ₁₀ alkylene-C(=O)-, -C3-C8 heterocyclo-C(=O)-, -C1C ₁₀ alkylene-(C ₃ -C ₈ heterocyclo)-C(=O)-, -(C3 -C8 heterocyclo)-C1-C ₁₀ alkylene-C(=O)-, -C1-C ₁₀ alkyleneNH-, C1-C ₁₀ heteroalkylene-NH-, -C3-C8 carbocyclo-NH-, -O-(C1- C ₈ alkyl)-MH-, -arylene-ΝΗ-, -C1-C ₁₀ alkylene-arylene-NΗ-, arylene-C1-C ₁₀ alkylene-NΗ-, -C1-C ₁₀ alkylene-(C ₃ -C ₈ carbocyclo)-NH-, - (C3-C8 carbocyclo)-C1-C ₁₀ alkylene-NH-, -C ₃ -C ₈ heterocyclo-NH-, -C1-C ₁₀ alkylene-(C ₃ -C ₈ heterocyclo^M ^, -(C ₃ -C ₈

- 48 046498 heterocyclo)-C1-C ₁₀ alkylene-NH-, -C1-C ₁₀ alkylene-S-, C ₁ -C _1o heteroalkylene-S-, -C ₃ -C ₈ carbocyclo-S-, -O(C1 -C ₈ alkyl)^-, -arylene-S-, -C1-C10 alkylene-arylene-S-, -arylene-C1-C ₁₀ alkylene-S-, -C1-C10 alkylene-(C ₃ C ₈ carbocyclo) ^-, -(C ₃ -C ₈ carbocyclo)-C1-C1 ₀ alkylene-S-, -C ₃ -C ₈ heterocyclo-S-, -C1-C10 alkylene-(C ₃ -C ₈ heterocyclo)^- or -(C ₃ -C ₈ heterocyclo)-C1-C1 ₀ alkylene-S-. Any of the R1 ⁷ substituents may be substituted or unsubstituted. In some aspects of R1 ⁷ the substituents are unsubstituted. In some aspects, R1 ⁷ substituents are optionally substituted. In some aspects, the R1 ⁷ group is optionally substituted with an alkaline group, for example -(CH2)xNH2, -(CH2)xNHR ^& -(CR2) _xNR ^a 2, where x is an integer ranging from 1-4, and each R ^a is independently selected from the group consisting of C1-6 alkyl and _C1-6 heloalkyl, or two R ^a groups are combined with the nitrogen atom to which they are bonded to form an azetidinyl, pyrrolidinyl or piperidinyl group. It should be clear, even when not explicitly stated, that p has a value between 1 and 14.

An exemplary extending group is a group of formula Xva, where R ¹⁷ is -C2-C5 alky'1ene-C(O)-, where the alkylene is optionally substituted with an alkali group, for example (CH2)xNH2, -(CH2)xNHR ^a and -( CH2) _x NR ^a 2, where x is an integer ranging from 1-4, and each R ^{a is} independently selected from the group consisting of C1-6 alkyl and C _1-6 heloalkyl, or two R ^a groups are combined with the nitrogen atom to which they are bonded to form an azetidinyl, pyrrolidinyl or piperidinyl group. Exemplary embodiments are as follows:

It will be understood that the substituted succinimide may exist in hydrolyzed form, as shown below

Exemplary extension groups prior to ligand conjugation include the following:

- 49 046498

It will be appreciated that the amino group of an extending group may be suitably protected with an amino protecting group during the synthesis, for example an acid labile protecting group (eg BOC).

Another illustrative broadening group is a group of formula XVb, where R ¹⁷ represents -(CH ₂ ) ₅ -

In another embodiment, the extending group is linked to the ligand group through a disulfide bond between the sulfur atom of the ligand and the sulfur atom of the extending group. A representative extension group of this embodiment is represented in square brackets of Formula XVI, wherein a wavy line indicates attachment in a ligand-drug conjugate or intermediates thereof, and R ¹⁷ has the meaning as defined above for Formula XVa and XVb.

In yet another embodiment, the reactive group of the extending group contains a reactive site that can form a bond with a primary or secondary amine group of the ligand. Examples of these reactive sites include, but are not limited to, activated esters such as succinimide esters, 4-nitrophenyl esters, pentafluorophenyl ethers, tetrafluorophenyl ethers, anhydrides, acid chlorides, sulfonyl chlorides, isocyanates and isothiocyanates. Representative Extension Groups of this embodiment are shown in square brackets of formulas XVIIa, XVIIb and XVIIc, where the wavy line indicates attachment in a ligand-drug conjugate or intermediates thereof, and R ¹⁷ has the meaning as defined above for formulas XVa and XVb.

L--C(O)NH—R ¹⁷ —;

XVIIa

C(O)r ¹⁷ —:

XVI lb

S

II'

L--CNH----R ¹⁷ - ^L -I XVIIc

In yet another embodiment, the reactive group of the Extending Group contains a reactive site that is capable of reacting with a (-CHO) modified carbohydrate group that may be present on the ligand. For example, a carbohydrate can be slightly oxidized using a reagent such as sodium periodate, and the resulting (-CHO) group of the oxidized carbohydrate can be condensed with an extending group that contains a functional group such as a hydrazide, an oxime, a primary or secondary amine, hydrazine, thiosemicarbazone , hydrazine carboxylate and aryl hydrazide, such as those described in Kaneko, T. et al. (1991) Bioconjugate

- 50 046498

Chem. 2:133-41. Representative extending groups of this embodiment are shown in square brackets of Formulas XVIIIa, XVIIIb and XVIIIc, where the wavy line indicates attachment in a ligand-drug conjugate or intermediates thereof, and R ¹⁷ has the meaning defined above for Formulas XVa and XVb.

In some embodiments of the present invention, it is desirable to increase the length of the extension group. Accordingly, the extension group may include additional components. For example, an expanding group may include the groups shown in the square brackets of the formulas XVa1,

wherein the wavy line indicates attachment to the rest of the ligand-drug conjugate or intermediates thereof;

R ¹⁷ has the meaning defined above, preferably R ¹⁷ is -C2-C5 alkyleneC(=O)-, where the alkylene is optionally substituted with an alkaline group, for example -(CH2)xNH2, -(CH2)xNHR ^a and -(CH2) xNR ^a 2, where x is an integer ranging from 1-4, and each R ^{a is} independently selected from the group consisting of C1-6 alkyl and C _1-6 heloalkyl, or two R ^a groups are combined with a nitrogen atom, to which they are linked to form an azetidinyl, pyrrolidinyl or piperidinyl group; And

R ¹³ represents -C1-C6 alkylene-, -C ₃ -C ₈ carbocyclo-, -arylene-, -C1-C10 heteroalkylene-, -C ₃ C ₈ heterocyclo-, -^-C^alkylene-arylene-, - arylene-C ₁ -C ₁₀ alkylene-, -C ₁ -C ₁₀ alkylene-(C ₃ -C ₈ carbocyclo)-, -(C ₃ -C ₈ carbocyclo)-C ₁ -C ₁₀ alkylene-, -^-^ alkylene-^-^ heterocyclo)- or -(C3-C8 heterocyclo)-C1-C10 alkylene-. In preferred embodiments, R ¹³ is -C1-C6 alkylene-.

In preferred aspects of the present invention, the expansion group has a mass of no more than about 1000 Da, no more than about 500 Da, no more than about 200 Da, from about 30, 50 or 100 to about 1000 Da, from about 30, 50 or 100 Da to about 500 Da, or from about 30, 50 or 100 Da to about 200 Da.

Optional branching group (A)

A branching moiety is included in ligand-drug conjugates in cases where it is desired to add additional drugs to the drug-linker moiety and ultimately to the ligand. The branching group is capable of forming a covalent bond with two to four parallel connecting units, two to four Drug Attachment groups, or two to four -X-D groups. Thus, the branching group allows the attachment of several

- 51 046498 groups in structures such as

in cases where m has a value greater than 1. Those skilled in the art will appreciate that the branching group is designed to allow branching in the linker. To act as a branching group, the branching group contains at least a first, second and third attachment site for attachment in a ligand-drug conjugate or intermediates thereof. In other words, the branching group must be at least trifunctional. In embodiments where m is 3 of 4, the branching group will contain four or five sites for covalent attachment in the ligand-drug conjugate or intermediates thereof. In some aspects, the branching group is a single group or contains two or more subunits (e.g., 2 to 10, preferably 2 to 5, e.g., 2, 3, 4, or 5) to provide the desired number of attachment sites where the branching group or its subunit is independently selected from natural or non-naturally occurring amino acids, amino alcohols, amino aldehydes or polyamines or combinations thereof. If necessary, to have the desired number of attachments, at least one of the amino acids, amino alcohols, amino aldehydes or polyamines will contain a functionalized side chain to provide attachment sites for the LP group and/or Z group and/or AD groups and/or X-D groups. In some aspects, one or more amino acids, amino alcohols, or amino aldehydes must be non-naturally occurring and must be modified to contain one or more functionalized side chains for attachment sites. Examples of functionalized amino acids, amino alcohols, or amino aldehydes include, for example, azide- or alkyne-functionalized amino acids, amino alcohols, or amino aldehydes (e.g., an amino acid, amino alcohol, or amino aldehyde modified to contain an azide group or alkyne group for attachment, using click chemistry).

Each amino acid, amino alcohol, aminoaldehyde or polyamine can be of natural or non-natural origin. Likewise, each amino acid can be a D- or L-isomer. In some embodiments, where a branching group may connect two parallel connecting units, two X-D Groups, or two drug attachment groups, such branching group, when assembled, has the formula shown below:

where the wavy line indicates two or three of the three attachment sites in the ligand-drug conjugate or intermediates thereof, and where R ¹¹⁰ represents

- 52 046498 *-sn ₂

*—ch ₂ ch ₂ coo-£*--(CH ₂ ) ₄ NHC(=N-NH)CH ₃ , *-ch ₂ o-|, *—(CH ₂ ) ₃ NHC(=NH)NH-^ *---(CH ₃ ) ₄ NHC(=NO)CH ₃ , *—ССОНз ' *-(CH ₂ ) ₃ NH-^*--(CH ₂ ) ₃ NHCONH-^- , *—ch ₂ conh- ^— *—ch ₂ coo-|> *---(CH ₂ ) ₃ NHC(=N-NH)CH ₃ , *--CH ₂ CH ₂ CH(OH)CH ₂ NH-^*—ch ₂ ch ₂ conh-^*—(CH ₂ ) ₄ NHC(=NH)NH-J>' *---(CH ₂ ) ₃ NHC(=NO)CH ₃ .

' *----(CH ₂ ) ₃ NHCH=N-NH-£- _{5 5} *—(ΟΗ^ΝΗ-β*—CH ₂ CH ₂ CH(O)CH ₂ NH ₂ , *---(CH ₂ ) ₃ NHCH=NO-^- _; *--(CH ₂ ) ₄ NHCONH-£*—(C(CH ₃ )(CH ₃ )S-<l *—(C(CH ₃ )(CH ₃ ) NH-^*—(CHs^s-^- ,

*-CH ₂ 4^ N or *—CH ₂

where the asterisk indicates attachment to the carbon labeled x and the wavy line indicates one of the three attachment sites of the branching group;

each R ¹⁰⁰ is independently selected from hydrogen or -C ₁ -C ₃ alkyl, preferably hydrogen or CH3,

Y is independently selected from N or CH, each Y' is independently selected from NH, O or S, the subscript c is independently an integer having a value from 1 to 10, preferably from 1 to 3.

In preferred embodiments, R ¹¹⁰ is other than *—CH2CH ₂ CH(O)CH ₂ NH2

In some embodiments, where the branching group is capable of binding to two parallel connecting units or two drug attachment groups, each branching group in the ligand-drug conjugate or intermediates thereof, once assembled, has the formula shown below

where the subscript n has a value from 1 to 4;

X ^b is selected from the group consisting of -O-, -NR-, -S- -C(=O)- and -S(=O)-; And

R ¹ and R ² are independently selected from the group consisting of H, C _1-3 alkyl, phenyl and a C ₂ -C 5 heterocycle (preferably H or C _1-3 alkyl), where the wavy line indicates covalent attachment of a branching group in the ligand conjugate -drug or its intermediate.

The amino acid, amino alcohol, amino aldehyde or branching group polyamine may optionally be substituted with optionally substituted C _1-20 heteroalkylene (preferably optionally substituted C _1-12 heteroalkylene), optionally substituted C _3-8 heterocyclo, optionally substituted C _6-14 arylene or optionally substituted C ₃ -C ₈ carbocyclo, as described herein. The optionally substituted heteroalkylene, heterocycle, arylene or carbocyclo must contain functional groups for attachment in the ligand-drug conjugate or intermediates thereof.

Optional substituents include (=O), -X, -R, -OR, -SR,, -NR ₂ , -NR ₃ , =NR, -CX ₃ , -CN, -OCN, -SCN, -N=C= O, -NCS, -NO, -NO2, =N ₂ , -N3, -NRC(=O)R, -C(=O)R, -C(=O)NR2, -SO3 ^- , -SO3H, - S(=O)2R -OS(=O)2OR, -S(=O)2NR, -S(=O)R, -OP(=O)(OR)2, -P(=O)(OR) 2, -PO ^- 3, -PO3H2, -ASO2H2, -C(=O)R, -C(=O)X, -C(=S)R, -CO2R, -CO2 ^- , -C(=S) OR, -C(=O)SR, -C(=S)SR, -C(=O)NR2, -C(=S)NR2 or -C(=NR)NR2, where each X is independently a halogen: -F, -Cl, -Br or -I; and each R independently represents -H, -C ₁ -C ₂₀ alkyl, -C ₆ -C ₂₀ aryl, -C ₃ -C ₁₄ heterocycle, protecting group or prodrug group.

- 53 046498

Preferred optional substituents are (=O), -X, -R, -OR, -SR and -NR2.

An example of a branching group is lysine, as shown below, where the wavy line and asterisk indicate the covalent bond in the ligand-drug conjugate or its intermediates

It will be appreciated that in the formulas for some of the intermediates presented herein, the optional branching group is capable of forming two to four covalent attachments to the -X-D groups, but is not yet attached to them. In such embodiments, the branching group will be in partially assembled form and, as such, will include two or more reactive functional groups that can react with groups present in the releasable complex groups of the -X-D portions of the conjugate. Particularly preferred reactive functional groups include sulfhydryl groups capable of forming disulfide bonds, or thioesters.

In preferred aspects of the present invention, the branching group has a mass of no more than about 1000 Da, no more than about 500 Da, no more than about 200 Da, from about 10, 50 or 100 Da to about 1000 Da, from about 10, 50 or 100 Da to about 500 Da, or from about 10, 50 or 100 Da to about 200 Da.

Drug Attachment Group (AD)

As with the branching group, a drug attachment group is included in ligand-drug conjugates in cases where it is desirable to add additional -XD groups (i.e., a releasable complex group covalently attached to the drug group) to the drug-linker portion and , ultimately to the ligand. The drug attachment group, depending on placement in the ligand-drug conjugate or its intermediates, will contain either two attachment sites or three attachment sites for binding to the components of the ligand-drug conjugate or its intermediates. Those skilled in the art will appreciate that a drug attachment group can be any group that serves to provide attachment of an additional -X-D group to the drug linker moiety and ultimately to the ligand group. In some embodiments, each drug attachment group is a single group or contains two or more subunits (e.g., 2 to 10, preferably 2 to 5, e.g., 2, 3, 4, or 5), wherein the drug attachment group or its subunit is independently selected from natural or non-naturally occurring amino acids, amino alcohols, aminoaldehydes, diamines or polyamines or combinations thereof. If desired, to have the required number of attachments, at least one of the amino acids, amino alcohols, amino aldehydes or polyamines will contain a functionalized side chain to provide attachment sites for the LP group, and/or Z group, and/or AD groups, and/or X-D groups. In some aspects, one or more amino acids, amino alcohols, or amino aldehydes must be non-naturally occurring and must be modified to contain one or more functionalized side chains for attachment sites to a releasable complex group. Examples of functionalized amino acids, amino alcohols, or amino aldehydes include, for example, azide- or alkyne-functionalized amino acids, amino alcohols, or amino aldehydes (e.g., an amino acid, amino alcohol, or amino aldehyde modified to contain an azide group or alkyne group for attachment, using click chemistry).

In some aspects, when the AD group contains three attachment sites, the AD group, in its assembled form, has the formula presented below:

- 54 046498

I _R 100 jvw Q

Vy'-.-y·/ where the wavy line indicates two or three of the three AD attachment sites in the ligand-drug conjugate or intermediates thereof, and where R ¹¹⁰ represents

where the asterisk indicates attachment to the carbon labeled x and the wavy line indicates one of the three attachment sites;

R ¹⁰⁰ is independently selected from hydrogen or -C ₁ -C ₃ alkyl, preferably hydrogen or CH3,

Y is independently selected from N or CH,

Y' is independently selected from NH, O or S and the subscript c is independently selected from 1 to 10, preferably 1 to 3.

In preferred aspects, R ¹¹⁰ is other than *—CH2CH ₂ CH(O)CH ₂ NH2

I σνχΓΌ I

In embodiments where the AD group contains two attachment sites (ie, a terminal AD group), one of the attachment sites shown above may be substituted with, for example, H, OH, or a C _1-3 unsubstituted alkyl group.

In some embodiments, where the AD group contains three attachment sites, AD

- 55 046498 group, in its assembled form, independently has the formula presented below

where the wavy line indicates the sites of attachment in the ligand-drug conjugate or intermediates thereof and where x, R is ¹⁰⁰ or greater, and where

R ¹¹¹ represents p-hydroxybenzyl,

-CH(OH)CH3, -CH2CH2SCH3, -CH2CONH2,

R ¹¹⁰ have values determined directly

-(CH2)3NHC(=NH)NH2, -(CH2)3NH2,

-(CH2)4NH2, (CH2)4NHCOCH3,

-(СН2)зКНСОНз,

-(CH2)4NHCHO, methyl, isopropyl, isobutyl, sec-butyl, -CH ₂ OH,

-CH2COOH, -CH2CH2CONH2, -CH2CH2COOH,

-(CH2)3NHCHO, -(CH2)4NHC(=NH)NH2,

-(CH2)3NHCONH2, -(CH2)4NHCONH2,

CH ₂ CH2CH(OH)CH ₂ NH2, 2-pyridylmethyl-, 3-pyridylmethyl-, 4-pyridylmethyl-

In some embodiments, where the AD group contains three attachment sites, the AD group consists of two or more amino acids. An example of such an amino AD group is cysteine alanine, as shown below, where the wavy line and asterisk indicate attachment in a ligand-drug conjugate or its intermediates

In some embodiments, an asterisk indicates covalent attachment to a releasable complex group.

In some embodiments, where the AD group contains two attachment sites, the AD group consists of two or more amino acids. An example of such an amino AD group is cysteine alanine, as shown below, where the wavy line and asterisk indicate attachment in a ligand-drug conjugate or its intermediates

In some embodiments, an asterisk indicates covalent attachment to a releasable complex group.

The amino acid, amino alcohol, amino aldehyde or polyamine of the AD group may optionally be replaced with optionally substituted C _1-20 heteroalkylene (preferably optionally substituted C 1 ₁₂ heteroalkylene), optionally substituted C _3-8 heterocyclo, optionally substituted C _6-14 arylene or optionally substituted C ₃ -C ₈ carbocyclo, as described in this application. The optionally substituted heteroalkylene, heterocycle, arylene or carbocyclo must contain functional groups for attachment in the ligand-drug conjugate or intermediates thereof. Optional substituents include (=O), -X, -R, -OR, -SR, -NR2, _-NR3 ,=NR, CX3, CN, OCN, SCN, N=C=O, NCS, NO, NO2, =N ₂ , N ₃ , NRC(=O)R, -C(=O)R, -C(=O)NR ₂ , SO ₃ ^- , SO ₃ H, S(=O) ₂ R, -OS( =O)2OR, -S(=O) ₂ NR, -S(=O)R, -OP(=O)(OR)2, -P(=O)(OR)2, PO ^- z, PO3H2, ASO2H2, C(=O)R, C(=O)X,

- 56 046498

C(=S)R, CO2R, CO ₂ -, C(=S)OR C(=O)SR, C(=S)SR C(=O)NR ₂ , C(=S)NR ₂ or C( =NR)NR ₂ where each X independently represents a halogen: -F, -Cl, -Br or -I; and each R independently represents -H, -C ₁ -C ₂₀ alkyl, -C ₆ -C ₂₀ aryl, -C ₃ -C ₁₄ heterocycle, protecting group or prodrug group. Preferred optional substituents are (=O), X, R, OR, SR and NR2.

The drug attachment group may be linear or branched and may be represented by formula B

Wherein BB ¹ is independently selected from an amino acid, an optionally substituted C _1-20 heteroalkylene (preferably an optionally substituted C _1-12 heteroalkylene), an optionally substituted C _3-8 heterocyclo, an optionally substituted C _6-14 arylene, or an optionally substituted C ₃ -C ₈ carbocyc lo;

and the subscript u is independently selected from 0 to 4; where the wavy line indicates the sites of covalent attachment in the ligand-drug conjugate or intermediate thereof. The optionally substituted heteroalkylene, heterocycle, arylene or carbocyclo must contain functional groups for linking between the BB subunits and in the ligand-drug conjugate or intermediates thereof.

In some embodiments, in at least one instance, BB ¹ is an amino acid such that the group can be defined as an amino group of drug attachment. The subscript u may have a value of 0, 1, 2, 3, or 4. In some aspects, BB ¹ is an amino acid and has a value of 0. In some embodiments, the AD Group includes no more than 2 optionally substituted C _1-20 heteroalkylenes, optionally substituted _C3-8 heterocyclo, optionally substituted _C6-14 arylene or optionally substituted C3-C8 carbocyclo. In some embodiments, the AD group includes no more than 1 optionally substituted _C1-20 heteroalkylene, optionally substituted _C3-8 heterocyclo, optionally substituted _C6-14 arylene, or optionally substituted C3-C8 carbocyclo. The optionally substituted heteroalkylene, heterocycle, arylene or carbocyclo must contain functional groups for linking between the BB subunits and in the ligand-drug conjugate or intermediates thereof

The drug-attachment amino acid may be an alpha, beta or gamma amino acid and may be of natural or non-natural origin. The amino acid may be a D or L isomer. Linking at the amino group of the drug attachment or to other components of the conjugate (or linker) can be accomplished, for example, through amino, carboxy or other functional groups. The optionally substituted heteroalkylene must contain functional groups for attachment in the ligand-drug conjugate or intermediates thereof. Methods for independent activation and interaction of functional groups are well known in the art.

In any of the embodiments provided herein, the drug-attachment group amino acid (including the drug-attachment amino group) may be independently selected from the D or L isomer of a thiol-containing amino acid. The thiol-containing amino acid may be, for example, cysteine, homocysteine or penicillamine.

In another embodiment, the amino acid that constitutes the drug-attachment group (including the amino-drug-attachment group) may be independently selected from the group consisting of L- or D-isomers of the following amino acids: alanine (including β-alanine), arginine, aspartic acid, asparagine, cysteine, histidine, glycine, glutamic acid, glutamine phenylalanine, lysine, leucine, methionine, serine, tyrosine, threonine, tryptophan, proline, ornithine, penicillamine, B-alanine, aminoalkynic acid, aminoalkandioic acid, heterocyclocarboxylic acid, citrulline, statin , diaminoalkanoic acid and their derivatives.

Preferred amino acids include cysteine, homocysteine, penicillamine, ornithine, lysine, serine, threonine, glutamine, alanine, aspartic acid, glutamic acid, selenocysteine, proline, glycine, isoleucine, leucine, methionine, valine and alanine.

It should be understood that in the formulas for some of the compounds described herein, such as compounds where the drug attachment group is capable of forming a covalent attachment to -X-D but is not yet bonded to -X-D, the drug attachment group will be in partial assembled form and, as such, will include a functional group that is capable of reacting with a group present in the releasable complex group. Particularly preferred reactive functional groups include sulfhydryl groups to form disulfide bonds or thioether bonds. In some aspects, the reactive sulfur atom will be protected by a protecting group. Thiol protecting groups

- 57 046498 or their use in chemical conjugation are well known in the art and include, for example, alkylthiol (eg tert-butylthiol, ethanethiol, 2-propanethiol, 2-pyridinethiol) protecting groups, aromatic thiol protecting groups (eg 2-pyridinethiol ) and acetyl protecting groups.

In preferred aspects of the present invention, the drug attachment group has a mass of no more than about 1000 Da, no more than about 500 Da, no more than about 200 Da, from about 10, 50 or 100 Da to about 1000 Da, from about 10.50 or 100 Da to about 500 Da, or from about 10, 50 or 100 Da to about 200 Da.

Released complex group (X)

The released complex group (-X-) links the drug group to the rest of the ligand-drug conjugate. The main function of the release complex group is to release free drug at the site targeted by the ligand. In this regard, the releasable complex group is capable of forming a cleavable bond with a drug moiety or contains a cleavable bond to release the drug (eg, as a result of antigen-mediated internalization). In preferred embodiments, the release mechanism for the releasable complex group is an enzymatic release mechanism or a disulfide elimination mechanism. The recognition site for the enzymatic release mechanism may be, for example, a peptide cleavage site or a sugar cleavage site (eg, a glucuronide cleavage site).

The releasable complex group may include 1 to 3 components, a cleavable group ( ^QCL ), an optional spacer group ( ^QSP ) and an optional covalent attachment group ( ^QCO ). The spacer group, when present, has the function of linking the cleavage group and the drug group. Accordingly, in embodiments where a spacer group is present, the spacer group will be directly linked to the drug group, and the cleavage group will be linked to the drug group through the spacer group. In embodiments where there is no spacer group, the cleavage group will be directly associated with the drug group.

Accordingly, the releasable complex group can be represented by the formula shown below, where Q ^CO represents a covalent attachment group, Q ^SP represents a spacer group, and Q ^CL represents a cleavable group. A covalent attachment group may or may not be present and a spacer group may or may not be present. The asterisk indicates the site of covalent attachment to the drug group, and the wavy line indicates the covalent attachment in the ligand-drug conjugate or its intermediate (to L ^P , A, or AD, as appropriate):

—Q ^co —Q ^cl —Q ^sp -*

In embodiments where a spacer group is absent and a covalent attachment group is present, -XD can be represented by formula XIX, where a wavy line adjacent to the covalent attachment group indicates covalent attachment to the rest of the linker (to L ^P , A or AD, depending depending on the situation).

—1—О ^с —Q ^cl —D < U и ^υ χιχ

In embodiments where there is no covalent attachment group and no spacer group, -XD can be represented by the formula XX, where the wavy line adjacent to the cleavable group indicates covalent attachment to the rest of the linker (to L ^P , A or AD, depending on situations):

—ί—Q ^cl -D « XX

In embodiments where a spacer group is present and a covalent attachment group is present, -X-D may be represented by formula XXI, where a wavy line adjacent to the covalent attachment group indicates covalent attachment to the remainder of the linker (to LP, A, or AD, depending on situations):

_^_Q ^C °_Q ^C1 —QSP-D

In embodiments where a spacer group is present and a covalent attachment group is absent, XD may be represented by formula XXII, where a wavy line adjacent to a cleavable group or spacer group indicates covalent attachment to the remainder of the linker ( ^LP , A, or AD, depending depending on the situation).

- 58 046498

Those skilled in the art will appreciate that any of the above definitions for -XD (Formula XIX-XXIV) can be used in any of the formulas and embodiments presented herein and in any of the specific embodiments thereof. Each X, D and each ^QCO , ^QCL or ^QSP group may be the same or different from each other.

In preferred aspects of the present invention, the releasable complex group has a mass of no more than about 5000 Da, no more than about 4000 Da, no more than about 3000 Da, no more than about 2000 Da, no more than about 1000 Da, no more than about 800 Yes or no more than about 500 Yes. In some aspects, the releasable complex group has a mass of about 100 Da or about 200 Da or about 300 Da to about 5000 Da or about 100 Da or about 200 Da or about 300 Da to about 4000 Da or about 100 Da or from about 200 Da or from about 300 Da to about 3000 Da, from about 100 Da or from about 200 Da or from about 300 Da to about 2000 Da, from about 100 Da or from about 200 Da or from about 300 Da to about 1000 Da, from about 100 Da or from about 200 Da or from about 300 Da to about 800 Da, or from about 100 Da or from about 200 Da or from about 300 Da to about 500 Da.

Those skilled in the art will recognize that components of linker intermediates or drug-linker compounds can be linked in the same manner as ligand-drug conjugates without a ligand group.

Cleavage group ( ^QCL )

The cleavable group is the only component of the releasable complex group that must be present. In some aspects, the cleavable group forms a cleavable bond with the drug group. In some aspects, the cleavable group forms a cleavable bond with the spacer group. In some aspects, the cleavable bond is within a cleavable group, but allows free drug to be released (eg, through a 1,6-elimination reaction after cleavage). Functional groups for the formation of a cleavable bond may include, for example, sulfhydryl groups for the formation of a disulfide bond, aldehyde, ketone or hydrazine groups for the formation of a hydrazone bond, hydroxylamine groups for the formation of an oxime bond, carboxyl or amino groups for the formation of peptide bonds, carboxyl or hydroxy groups for formation of an ester bond and sugar to form a glycosidic bond.

The nature of the fissile group can vary widely. For example, cleavable linkers include disulfide-containing linkers cleavable via disulfide exchange, acid-labile linkers cleavable at acidic pH, and linkers cleavable by hydrolases (eg, peptidases, esterases, and glucuronidases).

The structure and sequence of the cleaved group may be such that the group is cleaved by enzymes present at the target site. In other aspects, the cleaved group may be cleaved by other mechanisms. The cleavage group may include one or more cleavage sites.

In some embodiments, the cleavage group includes one amino acid or one or more amino acid sequences. The cleavable group may include, for example, a monopeptide, dipeptide, tripeptide, tetrapeptide, pentapeptide, hexapeptide, heptapeptide, octapeptide, nonapeptide, decapeptide, undecapeptide or dodecapeptide group.

Each amino acid of a cleavable group may be a natural or non-naturally occurring amino acid and/or a D or L isomer, provided, of course, that a cleavable bond is present. In some embodiments, the cleavage group includes only naturally occurring amino acids. In some embodiments, the cleavage group comprises 1 to 12 amino acids in contiguous sequence.

In some embodiments, each cleavable amino acid is independently selected from the group consisting of alanine, arginine, aspartic acid, asparagine, histidine, glycine, glutamic acid, glutamine, phenylalanine, lysine, leucine, serine, tyrosine, threonine, isoleucine, proline, tryptophan , valine, cysteine, methionine, selenocysteine, ornithine, penicillamine, β-alanine, aminoalkanoic acid, aminoalkynoic acid, aminoalkandioic acid, aminobenzoic acid, amino-heterocycloalkanoic acid, heterocyclo-carboxylic acid, citrulline, statin, diaminoalkanoic acid and their derivatives . In some embodiments, each amino acid is independently selected from the group consisting of alanine, arginine, aspartic acid, asparagine, histidine, glycine, glutamic acid, glutamine, phenylalanine, lysine, leucine, serine, tyrosine, threonine, isoleucine, proline, tryptophan, valine , cysteine, methionine and selenocysteine. In some embodiments, each amino acid is independently selected from the group consisting of alanine, arginine, aspartic acid, asparagine, histidine, glycine, glutamic acid, glutamine, phenylalanine, lysine, leucine, serine, tyrosine, threonine, isoleucine,

- 59 046498 proline, tryptophan and valine. In some embodiments, each amino acid is selected from proteinogenic or non-proteinogenic amino acids.

In another embodiment, each cleavable amino acid is independently selected from the group consisting of the following L-(naturally occurring) amino acids: alanine, arginine, aspartic acid, asparagine, histidine, glycine, glutamic acid, glutamine, phenylalanine, lysine, leucine, serine, tyrosine , threonine, isoleucine, tryptophan and valine.

In another embodiment, each cleavable amino acid is independently selected from the group consisting of the D-isomers of these naturally occurring amino acids: alanine, arginine, aspartic acid, asparagine, histidine, glycine, glutamic acid, glutamine, phenylalanine, lysine, leucine, serine, tyrosine, threonine, isoleucine, tryptophan and valine.

In some embodiments, the bond between the cleavable group and the drug group or spacer group may be enzymatically cleaved by one or more enzymes, including a tumor-associated protease, to release a drug group (-D), which in one embodiment is subject to in vivo protonation when released to provide drug (D).

Useful cleavable groups can be designed and optimized for their selectivity for enzymatic cleavage by a particular enzyme, such as a tumor-associated protease. In one embodiment, the binding (or linkage) between the cleavable group and the drug group or spacer group is such that cleavage of the bond is catalyzed by cathepsin B, C and D or the protease plasmin.

In some embodiments, the cleavage group may include only naturally occurring amino acids. In other embodiments, the cleavage group may include only non-naturally occurring amino acids. In some embodiments, the cleavable group may include a naturally occurring amino acid linked to a non-naturally occurring amino acid. In some embodiments, the cleavage group may include a naturally occurring amino acid linked to a D-isomer of the naturally occurring amino acid.

An exemplary cleavable group is the dipeptide -Val-Cit-, -Phe-Lys- or -ValAla.

In some embodiments, the cleavable group comprises a peptide and contains from 1 to 12 amino acids. In some such embodiments, the peptide is conjugated directly to a drug group and there is no spacer group. In some such embodiments, the peptide is a dipeptide.

In some embodiments, the cleavable group -CU- is represented by -(-AM-) _1-12 or (-AM-AM-) _1-6 , wherein AM in each case is independently selected from naturally occurring amino acids or non-naturally occurring amino acids. In one aspect, the AM is in each case independently selected from naturally occurring amino acids. Those skilled in the art will appreciate that amino acids are typically linked to a drug group or spacer group through functional groups present on the amino acid, such as its carboxylic acid or amino termini.

In other aspects, the cleavable group includes a sugar cleavage site. In some such embodiments, the cleavable group includes a sugar moiety (Su) linked via an oxygen glycosidic bond to the self-cleaving group. In such aspects, the self-cleavage group is considered part of the cleavage group, ^QCL .

A self-cleaving group is a trifunctional chemical group that is capable of covalently linking together three spaced chemical groups (i.e., a sugar group (via a glycosidic bond), a drug group (directly or indirectly through a Q ^SP spacer group), and an L ^P group , A Group or AD Group (directly or indirectly through the covalent attachment group Q ^CO ) A glycosidic bond is one that can be cleaved at the target site, initiating a sequence of self-cleavage reactions that leads to the release of the drug.

Accordingly, the cleavable group may include a sugar moiety (Su) linked via a glycosidic bond (-O'-) to a self-cleavable group (K) of the formula

Sugar

wherein the self-cleaving group K forms a covalent bond with the drug group (directly or indirectly through a spacer group) and a covalent bond with L ^P , AD or A (directly or indirectly through a covalent attachment group), as appropriate.

The fissile group can, for example, be represented by the formula

- 60 046498

where Su represents a sugar component, -O'- represents an oxygen glycosidic bond; each R is independently hydrogen, halogen, -CN or -NO2; and wherein the wavy line indicates attachment to L ^P , AD or A (either directly or indirectly through a covalent attachment group), and the asterisk indicates attachment to a drug group (either directly or indirectly through a spacer group, wherein the spacer group, when present, may be, for example, -C(=O)-).

In some such embodiments, the sugar cleavage site is recognized by betaglucuronidase and the cleavage group includes a glucuronide unit. The glucuronide unit may include a glucuronic acid linked via a glycosidic bond (-O'-) to a self-cleaving group (K) of the formula

Glucuronic acid

O' where the self-cleaving group K forms a covalent bond with the drug group (directly or indirectly through the spacer group) and a covalent bond with L ^P , AD or A (directly or indirectly through the covalent addition group), depending on the situation.

The glucuronide unit may, for example, be represented by the formula

where the wavy line indicates covalent attachment to L ^P , AD or A (either directly or indirectly through a covalent attachment group), and the asterisk indicates covalent attachment to a drug group (either directly or indirectly through a spacer group).

In some embodiments, the cleavage group includes a sugar cleavage site, X-D is represented by the following formula:

where Su represents a sugar component, -O'- represents an oxygen glycosidic bond; each R independently represents hydrogen or a halogen, -CN, -NO ₂ or other electron-withdrawing group, Q ^CO represents a Covalent Addition Group; where is the wavy one

- 61 046498 line indicates covalent attachment to the rest of the linker group (LP, A or AD, as appropriate).

When the cleavage group includes a glucuronide unit, -X-D can, for example, be represented by the following formula:

where the wavy bond indicates covalent attachment to the rest of the linker group (LP, A or AD, as appropriate); and Q ^CO represents a covalent addition group.

In some other embodiments, the cleavable group will itself include a sulfur atom that is capable of forming a bond with a sulfur atom of the spacer group or drug group to form a disulfide or hindered disulfide. The cleavage occurs between the two sulfur atoms of the disulfide. In some such embodiments, one of the sulfur atoms is removed from the drug group and, in the absence of any other release mechanism, the other sulfur atom remains attached to the drug group and becomes part of the drug group.

A wide variety of disulfide linkers are known in the art and can be adapted for use in the present invention, including, for example, those that can be formed using SATA (N-succinimidyl-acetylthioacetate), SPDP (N-succinimidyl-3(2- pyridyldithio)propionate), SPDB (Nsuccinimidyl-3-(2-pyridyldithio)butyrate), SMPT (Nsuccinimidyl-oxycarbonyl-alpha-methyl-alpha-(2-pyridyldithio)toluene) and SPP (N-succinimidyl 4-(2pyridyldithio) pentanoate). (See, for example, Thorpe et al., 1987, Cancer Res. 47:5924-5931; Wawrzynczak et al., In Immunoconjugates: Antibody Conjugates in Radioimagery and Therapy of Cancer (C.W. Vogel ed., Oxford U. Press, 1987. See (see also US Patent No. 4880935.)

In some embodiments, the cleavable group is pH-sensitive and will include, for example, an acid-labile linker that is hydrolyzable in the lysosome (for example, a hydrazone, semicarbazone, thiosemicarbazone, cis-aconitic acid amide, orthoester, acetal, or ketal group may be used). (See, for example, US Patent Nos. 5122368; 5824805; 5622929; Dubowchik and Walker, 1999, Pharm. Therapeutics 83:67-123; Neville et al., 1989, Biol. Chem. 264:14 653-14 661) . Such linkers are relatively stable under neutral pH conditions, such as in blood, but they are unstable at the lower pH of 5.5 or 5.0, the approximate pH conditions of the lysosome.

In some embodiments, the cleavable group can be conjugated directly to the drug group, and the cleavable group can be linked to the drug group through a cleavable peptide or disulfide bond.

Spacer group ( ^QSP )

The spacer group, when present, has the function of linking the drug group to the cleavable group. There are two general types of spacer group: self-cleaving and non-self-cleaving. A non-self-cleaving group is one in which part or all of the spacer group remains associated with the drug group after cleavage and can either further disintegrate or spontaneously degrade to provide free drug, or can itself become part of the drug group. Examples of a non-self-cleaving group include, but are not limited to, a glycine-glycine group and a glycine-alone group (both shown in Scheme A) (below). When a ligand-drug conjugate containing a glycine-glycine group or including a glycine group alone undergoes enzymatic cleavage via a tumor cell-associated protease, cancer cell-associated protease, or lymphocyte-associated protease, the glycine-glycine group

- 62 046498 drug or glycine-drug group is cleaved from the conjugate. In one embodiment, an independent hydrolysis reaction occurs in the target cell, cleaving the glycine-drug group bond and releasing the drug.

Scheme A

-<j-Gly—D -|“Gly—Gly—D enzymatic^jenzymatic I cleavage W cleavage W

Gly—D Gly--Gly—D hydrolysis hydrolysis drug drug

In one embodiment, the non-self-cleaving group is -Gly-Gly-. In another embodiment, the non-self-cleaving group is -Gly-.

In another embodiment, the Spacer Group includes a p-aminobenzyl alcohol (PAB) group (see Schemes B and C, below) wherein the phenylene moiety is replaced by a Qm group, wherein Q is -C1-C8 alkyl, -Q-(C1-C8 alkyl) or other electron-donating group, or -halogen, -nitro, -cyano or other electron-withdrawing group and m is an integer having a value from 04.

Alternatively, a conjugate containing a self-cleaving spacer group can release -D without the need for a separate hydrolysis step. In some aspects, the extender group includes a PAB group that is linked to a peptide cleavable group via the amino nitrogen atom of the PAB group and is linked directly to a drug moiety via a carbonate, carbamate or ester group. The PAB group and the adjacent carbonyl make up the spacer group. Without being bound by any particular theory or mechanism, Scheme B shows a possible drug release mechanism of a PAB group that is attached directly to -D via a carbamate or carbonate group, an idea supported by Toki et al, 2002, J Org. Chem. 67:1866-1872.

Scheme B

o enzymatic digestion

T

ABOUT

D

1,6- drug elimination wherein Q is -C1-C8 alkyl, -Q- (C1-C8 alkyl), halogen, -nitro or -cyano and m is an integer ranging from 0-4.

Without being bound by any particular theory or mechanism, Scheme C shows a possible drug release mechanism of a PAB group that is attached directly to -D via an ester or amine bond.

- 63 046498

Scheme C

enzymatic digestion

V

,6- drug elimination wherein Q is -C1-C8 alkyl, -O- ( _{C1 -C8} alkyl), halogen, -nitro or -cyano and m is an integer having a value of 0-4.

Without being bound by any particular theory or mechanism, Scheme D shows a possible drug release mechanism of the PAB group of a glucuronide unit that is attached directly to -D via a carbonyl.

Scheme D

Other examples of self-cleaving groups include aromatic compounds that are electronically similar to the PAB group, such as 2-aminoimidazole-5-methanol derivatives (see, for example, Hay et al., 1999, Bioorg. Med. Chem. Lett. 9: 2237) and ortho or para-aminobenzylacetals. Spacers that undergo cyclization by hydrolysis of the amide bond can be used, such as substituted and unsubstituted 4-aminobutyric acid amides (see, for example, Rodrigues et al., 1995, Chemistry Biology 2:223), suitably substituted with bicyclo[2.2. 1] and bicyclo[2.2.2] ring systems (see, e.g., Storm et al., 1972, J. Amer. Chem. Soc. 94:5815) and 2-aminophenylpropionic acid amides (see, e.g., Amsberry et al. al., 1990, J Org Chem 55:5867). Elimination of amine-containing drugs that are substituted at the α-position of glycine (see, eg, Kingsbury et al., 1984, J. Med. Chem. 27:1447) is also an example of a self-cleaving spacer useful in exemplary Conjugates.

In preferred embodiments of the present invention, the spacer group consists of 1, 2 or 3 self-cleaving or non-self-cleaving groups.

In preferred embodiments of the present invention, the spacer group has a mass of no more than about 1000 Da, no more than about 500 Da, no more than about 400 Da, no more than about 300 Da, or from about 10, 50 or 100 to about 1000 Da. from about 10, 50 or 100 to about 500 Da, from about 10, 50 or 100 Da to about 400 Da, from about 10, 50 or 100 Da to about 300 Da or from about 10, 50 or 100 Da to about 200 Da .

Covalent addition group ( ^QCO )

The covalent attachment group, when present, expands the scope of the releasable linker complex group to provide greater distance between the LP and the drug group. In this regard, the covalent attachment group contains a functional group that can form a bond with the functional group of the optional branching group A or LP or the drug attachment group AD at one end, and a functional group that can form a bond with the functional group of the cleavage group at the other ends. In some aspects, connections through non-conditionally cleavable bonds are illustrative.

Those skilled in the art will appreciate that a covalent attachment group can be any group or component that serves to provide attachment of a cleavable group to the rest of the molecule. In some aspects, the covalent group at

- 64 046498 compounds before assembly will contain two functional groups capable of forming a bond and attaching to components of the ligand-drug conjugate or its intermediate. One skilled in the art will appreciate that the preassembly covalent attachment group may contain more than two functional groups; however, for the purposes of the present invention, will attach through only two of the functional groups to the components of the ligand-drug conjugate or intermediate thereof. The covalent attachment group may consist of one or more (e.g., 1-10, preferably 1, 2, 3 or 4) natural or non-natural amino acids, amino alcohols, amino aldehydes, diamines, or a natural or non-natural amino acid, amino alcohol, aminoaldehyde or diamine. In some aspects, the covalent attachment group is a natural or non-naturally occurring amino acid, amino alcohol, amino aldehyde, or diamine. Examples of amino acids capable of acting as covalent attachment groups include β-alanine.

In some embodiments, the covalent addition group has the formula shown below:

where R ¹¹¹ represents p-hydroxybenzyl, methyl, isopropyl, isobutyl, sec-butyl, -CH2OH, -CH(OH)CH3, -CH2CH2SCH3, -CH2CONH2, -CH2COOH, -CH2CH2CONH2, -CH2CH2COOH, -(CH2)3NHC ( =NH)NH2, -(CH2)3NH2, -(СШзКНСОНз, -(СВДзКННСО, -(CH2)4NHC(=NH)NH2, (CH2)4NH2,

-(CH2)4NHCOCH3, -(CH2)4NHCHO, -(CH3KHCHO^, (CH2)4NHCONH2, -CH2CH2CH(OH)CH2NH2, 2pyridylmethyl-, 3-pyridylmethyl-, 4-pyridylmethyl-

each R ¹⁰⁰ is independently selected from hydrogen or -C ₁ -C ₃ alkyl, preferably hydrogen or CH3; and c is an integer independently selected from 1 to 10, preferably 1 to 3

A representative covalent addition group containing a carbonyl group for bonding to a cleavable group is as follows:

O —>—NH—R ¹³ —C—S— where R ¹³ represents -C1-C6 alkylene-, -C3-C8carbocyclo-, -arylene-, -C1-C1o heteroalkylene-, -C3-C8heterocyclo-, -CrC10alkylene -arylene-, -arylene-C1-C10alkylene-, -C1-C10alkylene-(C3-C8carbocyclo)-, -(C3-C8carbocyclo)-C1-C10alkylene-, -CrC10alkylene-(C3-C8 heterocyclo)- or -(C3 -C8 heterocyclo)-CrC10 alkylene. In preferred embodiments, R ¹³ is -C1-C ₆ alkylene.

A representative covalent addition group containing a carbonyl group for bonding to a cleavable group is as follows:

where R ¹³ represents -C1-C6 alkylene-, -C3-C8carbocyclo, -arylene-, -C1-C1o heteroalkylene, -C3-C8heterocyclo-, -CrC10alkylene-arylene-, -arylene-C1-C10alkylene-, -C1- C10alkylene-(C3-C8carbocyclo)-, -(C3-C8carbocyclo)-C1-C10alkylene, -CrC10alkylene-(C3-C8 heterocyclo)- or - (C3-C8 heterocyclo)-CrC10 alkylene-. In preferred embodiments, R ¹³ is -C1-C ₆ alkylene.

A representative covalent addition group containing an NH group for bonding to a cleavable group is as follows:

- 65 046498 o

C R ¹³ NH—%— where R ¹³ represents -C1-C6 alkylene, -C3-C8carbocyclo, -arylene, -C1-C10 heteroalkylene, -C3C8heterocyclo, -^-C^alkylene-arylene, -arylene-C1-C10alkylene , -C1-C10alkylene-(C3-C8carbocyclo)-, -(C3C^arbocyclo^^-C^alkylene-, -^-C^alkylene^C3-C3 heterocyclo)- or -(C3-C8 heterocyclo)^-^ alkylene. In preferred embodiments, R ¹³ is -C1-C ₆ alkylene.

A representative covalent addition group containing an NH group for bonding to a cleavable group is as follows:

where R ¹³ represents -C1-C6 alkylene-, -C ₃ -C ₈ carbocyclo, -arylene, -C ₁ -C _1o heteroalkylene-, -C ₃ -C ₈ heterocyclo, -^-C^alkylene-arylene-, -arylene-C1-C ₁₀ alkylene, -^-C^alkylene- (C ₃ -C ₈ carbocyclo)-, (C^C^arbocyclo^^-C^alkylene-, -^-C^alkylene-^-C heterocyclo)- or -(C ₃ -C ₈ heterocyclo)^alkylene-. In preferred embodiments, R ¹³ is -C1-C6 alkylene.

Specific embodiments of covalent addition groups include the following, wherein the wavy line adjacent to the nitrogen indicates covalent attachment to L ^P (or AD or A) and the wavy line adjacent to the carbonyl indicates covalent attachment to the cleavable group and m is an integer , having a value from 1 to 6, preferably from 2 to 6, more preferably from 2 to 4.

m or

O <> II ί»

-S—NH—CH ₂ -CH ₂ --C-£—

In some aspects, the covalent addition group is an optionally substituted C _1-8 heteroalkylene.

In some aspects, especially those where the covalent attachment group forms a bond with the sulfur atom of a parallel linking unit, branching group, or drug attachment group, the covalent attachment group will form a bond with that sulfur atom through the maleimide group of the covalent addition group.

Representative covalent attachment groups of this embodiment include the bracketed groups of Formulas XXIII and XXIV, wherein a wavy line indicates attachment to a cleavable group as defined herein and an asterisk indicates attachment to a sulfur atom of a parallel linking unit, branching group, or drug attachment group, and R ¹⁷ is -C1-C1o alkylene-, C1-C1o heteroalkylene-, -C3-C8 carbocyclo, -O-(C1-C8 alkyl), -arylene, -C1-C10 alkylene-arylene, -arylene-C1- C10 alkylene-, -C1-C10 alkylene-(C3-C8 carbocyclo)-, -(C3-C8 carbocyclo)^-^ alkylene-, -C3-C8 heterocyclo, -C1-C10 alkylene-(C3-C8 heterocyclo), -(C3-C8 heterocyclo)^-^ alkylene-, -C1-C10 alkylene-C(=())-, C1-C10 heteroalkylene-C(=O)-, -C3-C8 car6ocyclo-C(=O) -, -O-(C1-C8 alkyl)-C(=O)-, -arylene-C(=O), -C1-C10 alkylene-arylene-C(=O)-, -arylene-CrCw alkylene-C (=O)-, -C1-C10 alkylene-^-^ carbocyclo)-C(=O), - (C3-C8 carbocyclo)^-^ alkylene-C(=O)-, -C3-C8 heterocyclo-C (=O), -C1-C10 alkylene-(C3-C8 heterocyclo)-C(=O)-, -(C3-C8 heterocyclo)^-^ alkylene-C(=O)-, -Ci-Ci0 alkylene- NH-, C1-C10 heteroalkylene-NH-, -C3-C8 carbocyclo-NH-, -O- (C1-C8 alkyl)NH-, -arylene-NH-, -C1-C10 alkylene-arylene-NH-, arylene -C1-C10 alkylene-NH-, -C1-C10 alkylene-(C3-C8 carbocyclo)-J-, -(C3-C8 carbocyclo)^-^ alkylene-NH-, -C3-C8 heterocyclo-NH-, C1 -C10 alkylene-(C3-C8 heterocyclo) -NH-, -(C3-C8 heterocyclo)^-^ alkylene-NH-, -C1-C10 alkylene-S-, C1-C10 heteroalkylene-S-, -C3-C8 carbocyclo-S-, -O-(C1-C8 alkyl)^-, -arylene-S-, -C1-C10 alkylene-arylene-S-, -arylene-C1-C10 alkylene-S-, -C1-C10 alkylene - (C3-C8 carbocyclo)^-, -(C3-C8 carbocyclo)^-^ alkylene-S-, -C3-C8 heterocyclo-S-, -C1-C10 alkylene-(C3-C8 heterocyclo)^- or - (C3-C8 heterocyclo)^-^ alkylene-S-. The R ¹⁷ substituents may be optionally substituted. In some aspects, the R ¹⁷ substituents are unsubstituted. In some aspects, the R ¹⁷ group is optionally substituted with an alkaline group, for example, -(CH2) _xNH2 , -( _CH2 )xNHRa ^and -(CH2) ^xNRa2 , where x is an integer ranging from 1-4, and each R ^a is independently selected from the group consisting of C _1-6 alkyl and C _1-6 heloalkyl, or two R ^a groups are combined with the nitrogen atom to which they are bonded to form an azetidinyl, pyrrolidinyl or piperidinyl group.

- 66 046498

An exemplary covalent addition group is a group of formula XXIII, where R ¹⁷ is -C2-C5 αalkylene-C(=O)-, where the alkylene is optionally substituted with an alkali group, for example -(CH2)xNH2, (CH^NHR' ¹ and - (CH2) _x NR ^a 2, where x is an integer having a value of 1-4, and each R ^{a is} independently selected from the group consisting of C1-6 alkyl and C _1-6 heloalkyl, or two R ^a groups are combined with nitrogen atom to which they are bonded to form an azetidinyl, pyrrolidinyl, or piperidinyl group. Exemplary embodiments are as follows:

It will be appreciated that the substituted succinimide shown above may exist in a hydrolyzed form (ie, a water molecule is added through one rather than both carbonyl-nitrogen bonds).

It will be understood that the amino group of an extending group may be protected by an amino protecting group, for example, an acid labile protecting group (eg, BOC).

In preferred aspects of the present invention, the covalent addition group has a mass of no more than about 1000 Da, no more than about 500 Da, no more than about 400 Da, no more than about 300 Da, from about 10, 50 or 100 Da to about 500 Da. , from about 10, 50 or 100 Da to about 500 Da, from about 10, 50 or 100 Da to about 400 Da, from about 10, 50 or 100 Da to about 300 Da, or from about 10, 50 or 100 Da to about 200 Yes.

PEGylated scaffolds for conjugation

As will be appreciated by those skilled in the art, the size of the PEG group selected for use in the present invention will depend on the hydrophobicity of the drug and linker components of the drug-linker group prior to attachment of the PEG group thereto. Intermediates of formulas DD, X, XI or XII can act as PEGylated conjugation scaffolds that can be used to screen combinations of drugs and PEG moieties that provide ADCs having improved pharmacokinetic parameters and/or minimal aggregation. PEGylated conjugation scaffolds provide a platform to optimize the number of PEG subunits for a given drug-linker compound.

PEGylated conjugation scaffolds are specifically designed to allow parallel conjugation of different drug moieties to PEG to investigate the ability of PEG to mask hydrophobicity and improve pharmacokinetic parameters for a wide range of conventional linker drugs (i.e., linker drugs that do not contain a parallel linked PEG group in accordance with the present invention). It is preferable to select a PEG moiety of sufficient size that will mask the hydrophobicity of the linker drug, but is not so large as to adversely affect the ability of the ligand-drug conjugate to diffuse to the target site or penetrate target cells and release the drug.

In particularly preferred embodiments, traditional medicines

- 67 046498 linkers that can be used to optimize PEGs are those that contain a reactive group for conjugation with the thiol group of the antibody, for example maleimido-containing drug linkers and a protease-cleavable releasable complex group X. Accordingly, examples of X-D groups containing a releasable protease-cleavable complex group X for use with conjugation frameworks include the following, where d represents any Drug Group described herein:

In other particularly preferred embodiments, conventional drug linkers that can be used to optimize PEGs are those that contain a reactive group for conjugation to the thiol group of the antibody, such as maleimide-containing drug linkers, and a releasable complex group X cleavable by a glycosidase . Accordingly, examples of X-D groups containing a glycosidase-cleavable releasable complex group X for use with conjugation frameworks include the following, wherein D is any drug group described herein

In embodiments where the drug linkers that can be used to optimize the PEG are those that contain a reactive group for conjugation with a thiol-acceptor group, such as a maleimide group, the framework for conjugation will contain a protected thiol-containing moiety that, when unprotected, it is capable of covalent attachment to the thiol-accepting group of the linker drug. The protected thiol-containing moiety may be a component of a parallel linking unit (or branching or drug attachment group).

- 68 046498

An exemplary PEGylated conjugation scaffold has the formula DD, wherein the LP group includes an amino acid having the following formula:

where the subscript n represents an integer ranging from 1 to 4;

R ¹ and R ² are independently selected from the group consisting of H, C _1-3 alkyl, phenyl or C ₂ -C5 heterocycle (preferably hydrogen, methyl, ethyl or propyl); And

R ^pr represents a suitable thiol protecting group.

An exemplary PEGylated conjugation scaffold has the formula DD, wherein the L ^P group includes a protected cysteine, homocysteine, or penicillamine. D or L isomers of amino acids are suitable. An example of an amino acid for use as the L ^P group is cysteine, as shown below, with t-butylthio as a suitable protecting group.

Exemplary PEGylated scaffolds for conjugation into an appropriately protected ligand-linker intermediate include the following:

Other exemplary PEGylated scaffolds for conjugation into an appropriately protected ligand-linker intermediate include the following:

- 69 046498

Exemplary PEGylated conjugation scaffolds, after conjugation with drug-linker compounds, provide ligand-drug conjugates as shown below:

- 70 046498

Exemplary conjugation intermediate frameworks have the formula (CC), wherein the LP group includes an amino acid having the following formula:

- 71 046498

where the subscript n is an integer ranging from 1 to 4;

R ¹ and R ² are independently selected from the group consisting of H, C 1-3 alkyl, phenyl or C ₂ -C 5 heterocycle (preferably hydrogen, methyl, ethyl or propyl); And

R ^PR is a suitable thiol protecting group.

Exemplary PEGylated scaffold conjugate intermediates in suitably protected linker intermediates are shown below

- 72 046498

An exemplary PEGylated conjugate framework may have formula XI, wherein the LP group and the drug attachment group AD' each comprise an independently selected amino acid having the following formula:

where the subscript n is an integer ranging from 1 to 4;

R ¹ and R ² are independently selected from the group consisting of H, C _1-3 αlkyl, phenyl or C ₂ -C5 heterocycle (preferably hydrogen, methyl, ethyl or propyl); And

R ^pr represents a suitable thiol protecting group.

Exemplary PEGylated backbone conjugates of formula XI in an appropriately protected ligand-linker intermediate are shown below

Exemplary PEGylated scaffolds for conjugation following conjugation with drug linkers provide Formula II ligand-drug conjugates

- 73 046498

For PEGylated conjugation frameworks and intermediates, the extending group, Z or Z', PEG, ligand, protecting group R ^pr and subscript p have the meanings described in any of the embodiments presented herein. In illustrative aspects, the extending group is a maleimide-containing extending group described above. In illustrative embodiments, the PEG group contains from 6 to 72, from 10 to 72, or from 12 to 72 subunits, and the extender group is a maleimide-containing extender group described above and in any of the embodiments presented herein for XVa.

Accordingly, the present invention provides methods for selecting a PEG Group for use in a ligand-drug conjugate, which include the steps of (i) providing a framework for conjugation having formula (DD), wherein the parallel linking unit includes a thiol-protected cysteine, (ii) removing the protecting group from a thiol-protected cysteine to form an unprotected conjugation scaffold containing a free thiol, (iii) contacting the unprotected conjugation scaffold with a drug linker containing a functional group for covalently binding to the free thiol under conditions to form a ligand-drug conjugate. The methods may further include testing the pharmacokinetic parameters of the resulting ligand-drug conjugate (see, for example, Example 8 or 21). Also provided are ligand-drug conjugates prepared by such methods.

Also provided are methods for selecting a PEG moiety for use in a ligand-drug conjugate, which comprise the steps of (i) providing a conjugation scaffold having formula XI or XII, wherein the parallel linking unit and drug attachment group(s) include a thiol-protected cysteine, (ii) removing protecting group from a thiol-protected cysteine to form an unprotected conjugation scaffold containing a free thiol, (iii) contacting the unprotected drug conjugation scaffold with a linker containing a functional group for covalent binding to the free thiol under conditions to form a ligand-drug conjugate. The methods may further include testing the pharmacokinetic parameters of the resulting ligand-drug conjugate (see, for example, Example 21). Also provided are ligand-drug conjugates prepared by such methods.

Drug load

Typically, for ligand-drug conjugates of formulas I, II, III and AA, the number of drug-linker groups per ligand is indicated by the symbol p. When referring to individual ligand-drug conjugates in a population of such conjugates, p is an integer representing the number of drug-linker molecules per ligand. When referring to a composition containing multiple conjugates (ie, an LDC composition), p represents the average number of linker drugs per ligand and is more typically not an integer. In those cases in the experimental part describing LDC compositions consisting of

- 74 046498 antibody-drug conjugates (ADCs), where the drug load is specified as a specified number of drug/antibody groups (e.g. 8 loads, 16 loads or 32 loads), this value refers to the average drug load as well as drug loading of the predominant ADC in the composition, which depends on the number of reactive sites on the antibody that will react with the linker drug compound or, if appropriate, the intermediate ligand, followed by the introduction of -X-D. In a population of ligand-drug conjugates, there may be an average of 1 to 14 drug-linker groups per ligand, an average of about 6 to about 14, about 6 to about 12, about 6 to about 10, about 8 to about 14, about 8 to about 12, or about 8 to about 10 drug-linker groups per ligand. As an example of attachment to a ligand, one can mention attachment via thioester bonds. Examples of conjugation sites on a ligand are thiol interchain disulfide residues and/or residues embedded in the ligand, such as embedded cysteines. When referring to embodiments where the average drug loading is about 8, 10, 12, 14, 16, or 32, a value of 8, 10, 12, 14, 16, or 32 typically also refers to the drug loading of the predominant ligand-drug conjugate agent in the composition. Likewise, when referring to embodiments where an average of about 8 to about 14, about 8 to about 12, or about 8 to about 10 drug-linker groups per ligand are present, this value also typically refers to drug linker loading agent in the predominant ADC in the composition.

The average number of drug-linker groups per ligand group in the drug resulting from the conjugation reaction can be determined by traditional methods such as mass spectroscopy, enzyme-linked immunosorbent assay (ELISA), hydrophobic interaction chromatography (HIC) and HPLC. The quantitative distribution of ligand-linker-drug conjugates, indicated by the p index, can also be determined. In some cases, the separation, purification and characterization of homogeneous ligand-drug conjugates, where p is a specific value from the ligand-drug conjugate with other drug loads, can be accomplished by methods such as reverse phase HPLC or electrophoresis.

Compositions

The present invention provides compositions comprising any of the ligand-drug conjugates described herein. For example, the present invention provides compositions comprising a ligand-drug conjugate of formulas AA, I, II or III, and any of the specific embodiments thereof. The variables have the meanings defined herein in any of the embodiments.

When Formulas AA, I, II, or III do not represent individual LDC compounds but rather an LDC composition (i.e., a composition comprising a population of ligand-drug conjugates), the subscript p represents the average number of drug-linker molecules per Ligand molecule ( for example, an antibody molecule) in the composition. Similarly, when formulas DD, X, XI and XII do not represent individual ligand-linker intermediates, but a composition of ligand-linker intermediates (i.e., a composition including a population of ligand-linker intermediates), the subscript p represents the average the number of linker molecules per ligand molecule (eg, antibody) in the composition. It should be understood that the compositions may include a plurality (or population) of ligand-drug conjugates having varying numbers of drug-linker groups attached thereto (e.g., 1 to 14, 2 to 12, 4 to 12, 6 to 12, from 8 to 12), to reach the average p value. Alternatively, the composition may include a plurality (or population) of ligand-drug conjugates having the same or substantially the same number of linker drugs attached thereto (1 to 14) to achieve an average p value. The terms set or population are used synonymously in this context. There may be a small percentage of unconjugated antibody in the composition, which is also reflected in the average p value. For a composition comprising a population of ligand-drug conjugates of the present invention, there may be an average of 1 to 14 drug-linker groups per ligand, an average of about 6 to about 14, an average of about 6 to about 12, an average of about 6 to about 10, about 8 to about 14, about 8 to about 12, or about 8 to about 10 drug-linker groups per ligand. The use of PEG as disclosed herein is particularly suitable for ligand-drug conjugates having high drug loadings, for example, with an average drug loading of at least about 6, more preferably at least about 8 drug linker groups per ligand wherein each drug-linker group contains one or more -X-D components, preferably 1, 2 or 4. Accordingly, the compositions provided herein will preferably have an average drug-linker loading of at least about 8 drug-linker molecules. linker per ligand in the composition, and preferably contain about 8, 10, 12 or 16 to about 32 drug groups per ligand group.

- 75 046498

In some aspects, the compositions are pharmaceutical compositions comprising the ligand-drug conjugates described herein and a pharmaceutically acceptable carrier. For example, the present invention provides pharmaceutical compositions comprising a conjugate of formula I, II or III and any of the specific embodiments thereof. In some aspects, the pharmaceutical composition will be in liquid form. In some aspects, it will be a lyophilized powder.

The compositions, including pharmaceutical compositions, may be presented in purified form. In the context of the present invention, purified means that, when isolated, the isolate contains at least 95% and, in another aspect, at least 98% of the conjugate based on the weight of the isolate.

Pharmacokinetics

As stated above, the present inventors have discovered that the pharmacokinetic profile of certain ligand-drug conjugates can be significantly altered by the addition of a PEG moiety. In some cases, placing the PEG in a parallel orientation with the ligand moiety and the drug moiety reduces the plasma clearance of the ligand-drug conjugate and increases plasma drug content, which improves the desired pharmacological activity of such conjugates. Surprisingly, placing the PEG group in a sequential orientation with the ligand group and the drug group does not provide such improved pharmacokinetic effects and in some cases actually increases clearance and reduces the relative plasma concentration compared to the non-PEGylated counterpart. Prior to the present invention, when attempts were made to reduce hydrophobicity through PEGylation of a hydrophobic compound, the effects of PEG group orientation were not taken into account.

There are many ways to measure the pharmacokinetic parameters of a ligand-drug conjugate. One method involves determining the concentration of the ligand-drug conjugate, i.e. the amount of ligand-drug conjugate in a given volume of plasma or serum at a given point in time. Another method involves determining the rate at which the drug is eliminated from the body, i.e. the volume of plasma (or serum) cleared of the ligand-drug conjugate per unit time. The third method involves determining the area under the curve (AUC), i.e. integral of the concentration-time curve. Concentration, clearance, and AUC can be determined by plotting serum (or plasma) concentration of total antibody (μg/mL) on the ordinate (Y-axis) against time (days) on the x-axis (X-axis) after administration of the drug of interest. means, subject. For example, in one method, pharmacokinetic parameters are measured by injecting mice with a dose of (i) an unconjugated ligand, (ii) a ligand-drug conjugate of the present invention, and (iii) a comparative ligand-drug conjugate and collecting blood samples at various points. time after injection (eg, 1, 2, 3, 7, 14, 21, 28, 35, 42, 49, and 56 days) and serum separation. Serum (or plasma) concentrations can be measured by methods known in the art. For example, the concentration in serum (or plasma) can be measured by a sandwich ELISA to determine the total amount of ligand (eg, antibody) using a suitable detection mechanism. Serum (or plasma) concentrations for each animal can be analyzed using an appropriate program to obtain concentrations, drug elimination rates, and AUCs at specific time points. In another embodiment, pharmacokinetic data can be obtained using radiolabeled conjugates. For example, animals can be administered a radiolabeled ligand or ligand-drug conjugate and the plasma (or serum) concentration measured by liquid scintillation counting. In some embodiments, the animal model used is a rat model.

In some embodiments, the pharmacokinetic profile of a drug ligand conjugate of the present invention resembles that of its unconjugated ligand. Accordingly, this application provides ligand-drug conjugates having a clearance value within about 3 or about 2 times the clearance of an unconjugated ligand, and/or an AUC value of at least 25% or at least 30 % of the AUC value of the unconjugated ligand (for example, see Table 2).

In some embodiments, the pharmacokinetic profile of a ligand-drug conjugate of the present invention is improved compared to a comparative conjugate. Accordingly, this application provides ligand-drug conjugates having a better concentration value, clearance value and/or AUC value compared to a comparative conjugate (ie, one not containing a PEG group in a parallel orientation to the drug-linker group). By superior clearance, it is meant that the ligand-drug conjugate has a clearance that is at least 2 or at least 3 times better than the clearance of the reference Conjugate (e.g., 14.2 mL/day). kg compared to 48.6 or 57.8 ml/day/kg). By better AUC, it is meant that the Ligand-Drug Conjugate has an AUC that is at least

- 76 046498 at least 2x or at least 3x better than the AUC value of the comparative conjugate (for example, a value of 229.7 day*μg/ml compared to a value of 67 or 52 day*μg/ml).

The comparative conjugate may be the same or substantially the same conjugate not containing a PEG group, the same or substantially the same conjugate not containing a PEG group in a parallel orientation but containing a PEG group in a serial orientation with respect to a ligand group and a drug group. In some embodiments, the comparative conjugate is a conjugate comprising the same drug group and either no PEG group at all (i.e., the same or substantially the same conjugate without the PEG group) or containing a PEG group that is in sequential orientation with respect to the ligand group and the drug group (ie, the same or substantially the same conjugate containing a PEG group, but which is not in a parallel orientation). Typically, the ligand-drug conjugate and the reference conjugate have the same drug loading (the average number of drug groups per ligand group in the composition).

In the context of the present invention, the phrase the same or substantially the same conjugate, not containing a PEG group, generally refers to a conjugate comprising the same or substantially the same ligand group, drug group and linker group (for example, an extension group and releasable complex group) but does not contain a parallel connecting unit L ^P and a PEG group. With respect to the comparative conjugate not containing a PEG group that is most closely similar to the ligand-drug conjugate of the present invention, the comparative conjugate must include the same ligand group, drug group, release complex group, extender group and parallel connecting unit (and AD or A group, if necessary). The parallel connecting link, however, will not be attached to a PEG group, but will be terminated by a functional group such as, for example, an acetyl group (see, for example, compound 44 in the examples)

In the context of the present invention, the phrase is the same or substantially the same conjugate that does not contain a PEG group that is in a parallel orientation, but contains a PEG group that is in a serial orientation with respect to the ligand group and the drug group (i.e., such the same or substantially the same conjugate containing a PEG group but which is not in a parallel orientation) generally refers to a conjugate comprising the same or substantially the same ligand group, drug group, and linker group (e.g., extension group and a releasable complex group) but does not contain a parallel connecting unit L ^P and a PEG group attached thereto in a parallel configuration, and includes a PEG group in the linker in a serial orientation with the ligand group and the drug group.

The term substantially the same in this context means that there may be minor changes, but such changes occur primarily to simplify the chemical synthesis and coupling of the various components of the conjugate. For examples of comparative conjugates containing no PEG or containing a PEG group in a serial orientation, in comparison with a conjugate of the present invention containing a PEG group in a parallel orientation, see the Examples section.

Ligand-drug conjugates that exhibit significantly greater plasma clearance and correspondingly lower plasma drug content compared to the unconjugated ligand will be useful for the present invention because they can be modified as described herein to include a PEG moiety. Significantly greater plasma clearance compared to an unconjugated Ligand refers to a clearance value that is greater than 2x, greater than 3x, or greater than 4x the plasma clearance value for an unconjugated Ligand (see, for example, Table 2). A lower plasma concentration of a substance compared to an unconjugated Ligand refers to an AUC value that is 30% or less, 25% or less, or 20% or less of the AUC of the unconjugated Ligand (see, for example, Table 2).

In some embodiments, ligand-drug conjugates are provided having a clearance value within about 3x or within about 2x the clearance of the unconjugated Ligand, and/or an AUC value equal to at least 25% or at least 30% of the AUC value of the unconjugated Ligand .

In some embodiments, a drug that can be used as a drug moiety in the present invention is one that, when conjugated to a ligand in the form of a PEG-free ligand-drug conjugate or containing PEG in a sequential orientation, produces a ligand conjugate - a drug that demonstrates significantly greater plasma clearance and, accordingly, a lower content of the substance in the blood plasma compared to the unconjugated ligand. Significantly greater plasma clearance compared to an unconjugated ligand refers to a clearance value that is greater than 2x, greater than 3x, or greater than 4x the plasma clearance value for an unconjugated ligand (see, for example, Table 2). Lower levels of the substance in blood plasma

- 77 046498 compared to the provided unconjugated Ligand refers to an AUC value that is 30% or less, 25% or less, or 20% or less of the AUC of the unconjugated ligand (see, for example, Table 2).

Ligand-drug conjugates containing a hydrophobic drug group or hydrophobic drug-linker groups will be useful for the present invention because they can be modified, as described herein, to include a PEG group and can be seen to improve their pharmacokinetic parameters when application of the present invention.

In preferred embodiments, the ligand is an antibody.

Aggregation

The present inventors have also discovered that aggregation of certain Ligand-Drug Conjugates can be significantly reduced by adding a PEG group in a parallel orientation relative to the hydrophobic drug-linker group.

In some embodiments, a drug that can be used in the present invention is one that, when conjugated to a ligand in the form of a ligand-drug conjugate, does not contain PEG or contains PEG in a sequential orientation and contains an average of 4, 8, or 16 drug groups per ligand produces a ligand-drug conjugate that has aggregation levels, measured by SEC, of 4% or greater, 5% or greater, or 10% or greater.

The present invention provides populations of ligand-drug conjugates containing an average of 8 drug groups per ligand group or more, 10 drug groups per antibody or more, 12 drug groups per antibody or more, 16 drug groups per antibody or more, or 32 drug moieties per antibody having an aggregation level of about 1% or about 2% or about 3% (e.g., Formula 1 or II, where p is a value of 4 or 8, m is a value of 1, s is a value of zero, and t has the value zero; formula II, where p has the value 8, m has the value 2, s has the value 1, and t has the value zero)

In preferred aspects, the ligand group is an antibody.

Special Embodiments

Exemplary -X-D groups of the present invention include the following, wherein the wavy line indicates covalent attachment to the LP, A, or AD group, as appropriate

It will be appreciated that the substituted succinimide shown above may exist in a hydrolyzed form (ie, a water molecule is added through one rather than both carbonyl-nitrogen bonds).

Examples of drug-linker compounds of the present invention include compounds represented by the following structures:

or a pharmaceutically acceptable salt of such a compound, wherein the PEG group has the meaning defined in any of the embodiments presented herein and may be dispersed or non-dispersed, and n is an integer having a value from 6 to 72, from 8 before

- 79 046498

72, 10 to 72, 12 to 72, 12 to 38, 12 to 36, 6 to 24, or most preferably 8 to 24 or 12 to 24; R ^pr represents hydrogen or a protecting group, for example an acid-labile protecting group, for example BOC. In some embodiments, n is 8, 10, 12, or 24. For a population of ligand-drug conjugates (i.e., LDC compositions) prepared using a dispersed PEG moiety precursor that preferably has an average peak molecular weight corresponding to PEG a group containing from about 6 to 72, from 8 to 72, from 10 to 72, from 12 to 72, from 12 to 38, from 12 to 36, from 6 to 24, or most preferably from 8 to about 24 subunits, or from about 12 to about 38 subunits. When the PEG is non-dispersed, then each LDC of the LDC composition will typically contain a PEG group that has the same number of PEG subunits (-OCH2CH2), i.e. the same integer value n. The non-dispersed PEG group may, for example, have the structure

where R ²¹ represents a PEG blocking group, preferably -CH ₃ or -CH ₂ CH ₂ CO ₂ H, and n is an integer ranging from 8 to 12, 8 to 24 or 12 to 38.

Exemplary drug-linker compounds of the present invention that provide 2X drug loading are:

means include compounds represented by the following

and those structures in which mc-VC-PAB-D is replaced by mc-VA-PAB-D or mc-VA-D or any other X-D group;

where R ^PR represents hydrogen or a protecting group, for example an acid-labile protecting group, for example BOC;

mc-VC-PAB-D has the structure

mc-VA-PAB-D has the structure mc-VA-D has the structure

- 80 046498

and MDpr-PAB(gluc)-D has the structure wherein mc-VC-PAB-D, mc-VA-PAB-D, mc-VA-D and MDpr-PAB(gluc)-D are exemplary -X-D groups linked with a PEGylated scaffold, and where the wavy line indicates covalent binding of the succinimide ring of mc or MDpr to the sulfur atom of the PEGylated scaffold;

and PEG has the meaning defined in any of the embodiments presented herein and may be dispersed when describing a population of LDCs prepared using a dispersed PEG group precursor, wherein the dispersed PEG group precursor preferably has an average peak molecular weight corresponding to the PEG group, having n from about 8 to about 24 subunits or from about 12 to about 38 subunits, or is non-disperse (as determined by a PEG group having an integer n value, where each LDC of the LDC composition will contain a PEG group that has the same integer value n). In some embodiments, the non-dispersed PEG group has the structure

where R ²¹ represents a PEG blocking group, preferably -CH ₃ or -CH ₂ CH ₂ CO ₂ H, the wavy line indicates covalent binding of the PEG group to the PEGylated backbone, and n is an integer ranging from 8 to 24 or 12 to 38.

In some embodiments, the mc group in mc-VC-PAB-D, mc-VA-D, and mc-VA-PAB-D, where the mc group has the structure

where the wavy line to the succinimide group indicates covalent binding to the PEGylated backbone and the wavy line to the carbonyl indicates covalent binding to the rest of the -XD, in any of the above structures where such an mc group is present, replace the MDpr group with the structure

where R ^PR represents hydrogen or a protecting group to give MDpr-VC-PAB-D, MDprVA-D and MDpr-VA-PAB-D, which are additional exemplary -XD groups.

It will be appreciated that the substituted succinimide in MDpr in any of the MDpr-containing -X-D groups may exist in a hydrolyzed form (ie, a water molecule is added through one rather than both carbonyl-nitrogen bonds). The -X-D group including mc may also have a succinimide ring in hydrolyzed form.

Other exemplary drug-linker compounds of the present invention that provide 2X drug loading include the following:

- 81 046498

where mc-VA-D, mc-VC-PABA-D, mc-VA-PABA-D and MDpr-PAB(gluc)-D are exemplary XD groups as described for the 2X drug-loaded structures presented above, and where PEG _A and PEGV, independently selected, have the meaning defined in any of the embodiments for the PEG moieties provided herein, and may be dispersed when applied to a population of ligand-drug conjugates (i.e., LDC compositions), prepared using a dispersed PEG group precursor, wherein the dispersed PEG group precursor preferably has an average peak molecular weight corresponding to a PEG group having n from about 8 to about 24 subunits or from about 12 to about 38 subunits, or the PEG is non-disperse (t i.e. a PEG group containing a discrete number of PEG subunits, determined by an integer value, thus each LDC of an LDC composition consisting of such ADCs will contain a PEG group that has the same integer value n). In some embodiments, PEG _A is a non-dispersed PEG group having the structure

R ²¹ and/or PEGV is a non-dispersed PEG group having the structure

Where each R ²¹ is an independently selected PEG Blocking group, in each case n, independently selected, is an integer ranging from 8 to 24 or 12 to 38. In a preferred embodiment, one R ²¹ is -CH ₃ , and the other is -CH ₂ CH ₂ CO ₂ H. In some embodiments, the mc group, which has the structure

in any of the structures above where such a group is present, replace MDpr with a group that has the structure

where R ^pr represents hydrogen or a protecting group, yielding MDpr-VC-PAB-D, MDprVA-D and MDpr-VA-PAB-D as-XD,

- 82 046498

In other embodiments, the MDpr group in the structure above where such a group is present is replaced with an mc group to produce mc-PAB(gluc)D as -X-D.

It will be appreciated that the substituted succinimide in MDpr in any of the MDpr-containing -X-D groups may exist in a hydrolyzed form (ie, a water molecule is added through one rather than both carbonyl-nitrogen bonds). The -X-D group including mc may also have a succinimide ring in hydrolyzed form.

Exemplary drug-linker compounds of the present invention that provide 4X drug loading include the following:

where mc-VC-PAB-D is defined for the structures described above with 2X drug loading; and PEG has the meaning defined in any of the embodiments presented herein, and may be particulate when it refers to a population of ligand-drug conjugates (i.e., LDC compositions) prepared using a particulate PEG group precursor, wherein the particulate precursor The PEG group preferably has an average peak molecular weight corresponding to a PEG group having n from about 8 to about 24 subunits or from about 12 to about 38 subunits, or is non-disperse (i.e., a PEG group containing a discrete number of PEG subunits determined by an integer value such that each LDC of the LDC composition consisting of such ADCs contains a PEG group that has the same integer value n). In some embodiments, the non-dispersed PEG group has the structure

where R ²¹ represents a PEG blocking group, the wavy line indicates covalent binding to the PEGylated backbone, and n is an integer ranging from 8 to 24 or 12 to 38. Preferably, R ²¹ is -CH ₃ or -CH2CH2CO2H.

In some embodiments, mc-VC-PAB-D as the -X-D group replaces any of the -X-D groups described herein, including MDpr-VC-PAB-D, mc-VA-PAB-D, and MDpr-VA- PABD.

It will be appreciated that the substituted succinimide in MDpr in any of the MDpr-containing groups X-D may exist in a hydrolyzed form (ie, a water molecule is added through one rather than both carbonyl-nitrogen bonds). The -X-D group including mc may also have a succinimide ring in hydrolyzed form.

Other exemplary drug-linker compounds of the present invention that provide 4X drug loading include the following:

- 83 046498

where MDpr-PAB(gluc)-D has the value determined for the above structures with 2X drug loading; and PEG has the meaning defined in any of the embodiments presented herein and may be particulate when it refers to a population of ligand-drug conjugates (i.e., LDC compositions) prepared using a particulate precursor group PEG, wherein the dispersed PEG group precursor preferably has an average peak molecular weight corresponding to a PEG group having n from about 8 to about 24 subunits or from about 12 to about 38 subunits, or PEG _A is non-disperse (i.e., a PEG group containing a discrete the number of PEG subunits, determined by an integer value, such that each LDC of the LDC composition consisting of such ADCs contains a PEG group that has the same integer value n). In some embodiments, the non-dispersed PEG group has the structure

where R ²¹ is a PEG blocking group, the wavy line indicates covalent binding to the PEGylated backbone, and n is an integer ranging from 8 to 24 or 12 to 38. Preferably, R ²¹ is -CH ₃ or -CH ₂ CH ₂ CO ₂ H.

In some embodiments, the MDpr-PAB(gluc)-D as -X-D is replaced by a mcPAB(gluc)-D group.

It will be appreciated that the substituted succinimide in MDpr in any of the MDpr-containing groups X-D may exist in a hydrolyzed form (ie, a water molecule is added through one rather than both carbonyl-nitrogen bonds). The -X-D group including mc may also have a succinimide ring in hydrolyzed form.

Exemplary ligand-drug conjugates of the present invention include those represented by the following structures:

- 84 046498

- 85 046498

or a pharmaceutically acceptable salt of such a compound, wherein p is an integer having a value from 1 to 14, preferably 2 to 12, 6 to 12, 8 to 12 or 8 to 10, AB is an antibody, preferably a monoclonal antibody , D is a drug group, and n is an integer having a value of 6 to 72, 8 to 72, 10 to 72, 12 to 72, 12 to 36, or 38, 6 to 24, or most preferably from 8 to 24. PEG has the meaning defined in any of the embodiments presented in this application for PEG groups. It should be appreciated that Ab-substituted succinimide may exist in a hydrolyzed form (i.e., a water molecule is added through one rather than both carbonyl-nitrogen bonds), especially for those antibody-drug conjugates that incorporate groups , such as

where the wavy line indicates covalent binding to the rest of the drug-ligand group of the antibody-drug conjugate.

It should be understood that the representative structures presented above can also represent compositions, in which case p represents the average number of drug-linker molecules per ligand in the composition. In such embodiments, p is typically not an integer value and may range from 1 to 14, preferably 2 to 12, 6 to 12, 8 to 12, or 8 to 10.

Exemplary ligand-drug conjugates of the present invention that provide 2X drug loading include conjugates represented by the following structures:

and those structures wherein the -X-D group of mc-VC-PAB-D is replaced by any of the -X-D groups described herein, including mc-VA-PAB-D and MDpr-VA-PAB-D

- 86 046498

or a pharmaceutically acceptable salt of such a compound, wherein p is an integer having a value of 1 to 14, preferably 2 to 12, 6 to 12, 8 to 12 or 8 to 10, Ab is an antibody, preferably a monoclonal antibody , D is a drug group, and n is an integer having a value of 6 to 72, 8 to 72, 10 to 72, 12 to 72, 12 to 36 or 38, 6 to 24, or most preferably from 8 to 24. PEG has the meaning defined in any of the embodiments presented in this application for PEG groups. It will be appreciated that the substituted succinimide linked to Ab or S may exist in a hydrolyzed form (ie, a water molecule is added through one rather than both carbonyl-nitrogen bonds).

It will be appreciated that the succinimide in the Ab-substituted MDpr group or the -X-D group may exist in a hydrolyzed form (ie, a water molecule is added through one rather than both carbonyl-nitrogen bonds). The succinimide in the mc group substituted by Ab or in the -X-D group may also exist in hydrolyzed form.

In any of the embodiments described above, drug group D may be MMAE, shown below, where the wavy line indicates the site of attachment to the rest of the drug-linker structure.

In some preferred aspects, including those where D is MMAE, p is 6, 7, 8, 9, 10, 11, or 12. In some embodiments, including those where D is MMAE, the antibody is conjugated to a linker via the sulfur atom of the cysteine residue of an antibody. The cysteine residue may be of natural or non-natural origin. For example, in some aspects, the cysteine will be derived from an interchain disulfide. In other aspects, the cysteine residue will be derived from an inserted cysteine (eg, a cysteine inserted at position 239). In some aspects, the antibody will attach to drug-linker groups through its interchain disulfides and through embedded cysteines.

In any of the embodiments described above, drug group D may be MMAF, shown below, where the wavy line indicates the site of attachment to the rest of the drug-linker structure.

In any of the embodiments described above, drug group D may be a camptothecin compound, exemplifying camptothecin itself, as shown below, where the wavy line indicates the site of attachment to the rest of the drug-linker structure:

- 87 046498

In any of the embodiments described above, drug group D may be a vinca alkaloid compound exemplifying vinblastine hydrazide, as shown below, where the wavy line indicates the site of attachment to the rest of the drug-linker structure:

In any of the embodiments described above, drug group D may be an anthracycline compound, illustrated below, where the wavy line indicates the site of attachment to the rest of the drug-linker structure:

Examples of PEGylated scaffolds in thiol-protected linker intermediates and corresponding ligand-linker compounds of the present invention include the following:

- 88 046498

or a pharmaceutically acceptable salt of such a compound, wherein n is from 2 to 72, preferably from 4 to 72 or from 8 to 72 or from 8 to 24;

p has a value from 1 to 14, preferably about 2 to about 12; And

The Ab is an antibody, preferably a monoclonal antibody.

As will be appreciated, in the formulas presented above, ligand-substituted succinimides may exist in their hydrolyzed form (ie, a water molecule is added through one rather than both C-N bonds of the succinimide). Additionally, in any of the embodiments described above, the tert-butylthiol protecting group may be replaced by another suitable thiol protecting group.

Exemplary multiplex PEGylated scaffolds as linker intermediates and corresponding ligand-linker compounds of the present invention include the following:

- 89 046498

- 90 046498

where the wavy line indicates covalent attachment to the ligand group, R ²¹ is independently selected from PEG blocking groups, preferably methyl or 3-propionic acid, and n is independently from 2 to 72, preferably from 4 to 72, or from 8 to 72, or 8 to 24, with 24 being more preferred. The thiol protecting group may be replaced by another suitable thiol protecting group.

In some preferred embodiments, p is 6, 7, 8, 9, 10, 11, or 12. In some embodiments, the antibody is conjugated to a linker via the sulfur atom of the antibody's cysteine residue. The cysteine residue may be of natural or non-natural origin. On

- 91 046498 example, in some embodiments, the cysteine will be derived from an interchain disulfide. In other embodiments, the cysteine residue will be derived from an inserted cysteine (eg, a cysteine incorporated at position 239). In some embodiments, the antibody will attach to drug-linker groups through its interchain disulfides and through embedded cysteines.

In some aspects of the present invention, the ligand-drug conjugates contain no more than 50, no more than 45, no more than 40, no more than 35, no more than 30, or no more than 25 intermediate atoms between the ligand group and the drug group . In some aspects of the present invention, the ligand-drug conjugates contain no more than 40, no more than 35, no more than 30, or no more than 25 intermediate atoms between the ligand group and the cleavable group.

In some embodiments, in ligand-drug conjugates, fewer intermediate atoms are present between the ligand and the drug group than the number of atoms in the PEG group. In some embodiments, in ligand-drug conjugates, fewer intermediate atoms are present between the ligand and the cleavable group than the number of atoms in the PEG group.

In some embodiments, in ligand-drug conjugates, fewer intermediate atoms are present between the ligand and the drug group than the number of intermediate atoms between the distal end of the PEG group and the parallel connecting link. In some embodiments, in ligand-drug conjugates, fewer intermediate atoms are present between the ligand and the cleavable group than the number of intermediate atoms between the distal end of the PEG group and the parallel connecting link.

In preferred embodiments of the present invention, the drug is preferably an auristatin (eg, MMAE or auristatin having comparable or higher hydrophobicity than MMAE), the released complex group includes a glucuronide unit cleaved by beta-glucuronidase; and the PEG group includes at least 6, at least 8, at least 10 or at least 12 subunits, but not more than 72 subunits, preferably not more than 36 or 24 subunits. In preferred aspects, the PEG group will include from about 8 to about 24 subunits, most preferably about 12 subunits. Other components of the ligand-drug conjugate or intermediates thereof may be as described in any of the embodiments presented herein.

Preferred compositions of the present invention include a population of ligand-drug conjugates, wherein the ligand moiety is an antibody (e.g., an intact antibody), the drug moiety is an auristatin or a non-austatin (preferably an auristatin, e.g., MMAE or an auristatin having comparable or greater hydrophobicity than MMAE), the released complex group includes a glucuronide unit that is cleaved by beta-glucuronidase; The PEG group includes at least 6, at least 8, at least 10 or at least 12 subunits, but not more than 72 subunits, preferably not more than 36 or 24 subunits; and the average number of drug-linker groups per antibody in the composition is at least 6 or at least about 8. In preferred aspects, the PEG group will include from about 8 to about 24 subunits, most preferably about 12 subunits. Other components of the ligand-drug conjugate may be as described in any of the embodiments presented herein.

Methods of application

Cancer treatment

Ligand-drug conjugates are useful for inhibiting the proliferation of a tumor cell or cancer cell by inducing apoptosis in the tumor or cancer cell, or for treating cancer in a patient. Ligand-drug conjugates can accordingly be used in a variety of settings for the treatment of cancer. Ligand-drug conjugates can be used to deliver a drug to a tumor cell or cancer cell. Without being bound by theory, in one embodiment, the ligand moiety of a ligand-drug conjugate binds to or associates with a cancer cell- or tumor cell-associated antigen, and the ligand-drug conjugate can be taken up (internalized) into the tumor cell or cancer cell via receptor-mediated endocytosis or other internalization mechanism. The antigen may be attached to a tumor cell or cancer cell or may be an extracellular matrix protein associated with a tumor cell or cancer cell. Once inside the cell, through a cleavage mechanism, the drug is released inside the cell. In an alternative embodiment, the drug or drug group is cleaved from the ligand-drug conjugate outside the tumor cell or cancer cell, and the drug or drug group then enters the cell.

- 92 046498

In one embodiment, the ligand group binds to a tumor cell or cancer cell.

In another embodiment, the ligand group binds to a tumor cell or cancer cell antigen that is located on the surface of the tumor cell or cancer cell.

In another embodiment, the ligand group binds to a tumor cell or cancer cell antigen, which is an extracellular matrix protein associated with a tumor cell or cancer cell.

The specificity of the ligand group for a particular tumor cell or cancer cell can be important in determining which types of tumor or cancer are most effectively treated. For example, ligand-drug conjugates that target a cancer cell antigen present in hematopoietic cancers may be useful for the treatment of hematologic malignancies (eg, anti-CD30, anti-CD70, anti-CD19, anti-CD33 ligand binding group (eg, antibody) may be useful for the treatment of hematological malignancies). Ligand-drug conjugates that target a cancer cell antigen present on solid tumors may be useful for treating such solid tumors.

Types of cancer that can be treated with a ligand-drug conjugate include, but are not limited to, hematopoietic cancers, such as, for example, lymphomas (Hodgkin's lymphoma and non-Hodgkin's lymphomas) and leukemias and solid tumors. Examples of hematopoietic cancers include follicular lymphoma, anaplastic large cell lymphoma, mantle cell lymphoma, acute myeloid leukemia, chronic myelocytic leukemia, chronic lymphocytic leukemia, diffuse large B-cell lymphoma, and multiple myeloma. Examples of solid tumors include fibrosarcoma, myxosarcoma, liposarcoma, chondrosarcoma, osteogenic sarcoma, chordoma, angiosarcoma, endotheliosarcoma, lymphangiosarcoma, lymphangioendothelial sarcoma, synovioma, mesothelioma, Ewing's tumor, leiomyosarcoma, rhabdomyosarcoma, colon cancer, colorectal cancer, kidney cancer, pancreatic cancer, bone cancer, breast cancer, ovarian cancer, prostate cancer, esophageal cancer, stomach cancer, oral cancer, nasal cancer, throat cancer, squamous cell carcinoma, basal cell carcinoma, adenocarcinoma, sweat gland carcinoma, sebaceous gland carcinoma, papillary carcinoma, papillary adenocarcinoma, cystadenocarcinoma, medullary carcinoma, bronchial carcinoma, renal cell carcinoma, hepatoma, bile duct carcinoma, choriocarcinoma, seminoma, embryonal carcinoma, Wilm's tumor, cervical cancer, uterine cancer, testicular cancer, small cell lung carcinoma, bladder carcinoma, lung cancer, epithelial carcinoma, glioma, glioblastoma multiforme, astrocytoma, medulloblastoma, craniopharyngioma, ependymoma, pinealoma, hemangioblastoma, acoustic neuroma, oligodendroglioma, meningioma, skin cancer, melanoma, neuroblastoma and retinoblastoma.

Multimodal cancer therapy

Cancer, including, but not limited to, tumor, metastasis, or other disease or disorder characterized by uncontrolled cellular growth, can be treated or inhibited by administration of a ligand-drug conjugate.

In other embodiments, methods of treating cancer are provided, comprising administering to a patient in need thereof an effective amount of a ligand-drug conjugate and a chemotherapeutic agent. In one embodiment, the chemotherapeutic agent is one that the cancer has not been shown to be resistant to when used for treatment. In another embodiment, the chemotherapeutic agent is one that the cancer has been found to be resistant to when used for treatment. Ligand-drug conjugates can be administered to a patient who has also undergone surgery to treat cancer.

In some embodiments, the patient also receives additional treatment, such as radiation therapy. In a specific embodiment, the ligand-drug conjugate is administered simultaneously with a chemotherapeutic agent or radiation therapy. In another special embodiment, a chemotherapeutic agent or radiation therapy is administered before or after administration of the ligand-drug conjugate.

The chemotherapy agent can be administered over several sessions. Any one or combination of chemotherapeutic agents, such as standard of care chemotherapeutic agent(s), may be administered.

In addition, methods of treating cancer with a ligand-drug conjugate are provided as an alternative to chemotherapy or radiation therapy when chemotherapy or radiation therapy has proven or may prove to be too toxic, such as causing unacceptable or intolerable side effects, for the subject receiving the treatment. Optionally, other cancer treatment, such as surgery, radiation therapy, or chemotherapy, may be used for the patient receiving treatment, depending on which treatment is found acceptable or tolerable.

- 93 046498

Treatment of autoimmune diseases

Ligand-drug conjugates are useful for killing or inhibiting the replication of a cell that causes an autoimmune disease, or for treating an autoimmune disease. Ligand-drug conjugates can suitably be used in various environments to treat an autoimmune disease in a patient. Ligand-drug conjugates can be used to deliver a drug to a target cell. Without being bound by theory, in one embodiment, a ligand-drug conjugate binds to an antigen on the surface of a target cell and the ligand-drug conjugate is then taken up into the target cell via receptor-mediated endocytosis. Once inside the cell, the linker group is cleaved, resulting in the release of the drug or drug group. the released drug is then free to migrate into the cytosol and induce cytotoxic or cytostatic activities. In an alternative embodiment, the drug is cleaved from the ligand-drug conjugate outside the target cell and the drug or drug group then enters the cell.

In one embodiment, the ligand moiety binds to an autoimmune antigen. In one aspect, the antigen is on the surface of a cell involved in an autoimmune condition.

In another embodiment, the ligand group binds to an autoimmune antigen that is located on the surface of a cell.

In one embodiment, the ligand moiety binds to activated lymphocytes that are associated with an autoimmune disease state.

In yet another embodiment, the ligand-drug conjugate kills or inhibits the proliferation of cells that produce an autoimmune antibody associated with a particular autoimmune disease.

Specific types of autoimmune diseases that can be treated with ligand-drug conjugates include, but are not limited to, Th2 lymphocyte-related disorders (eg, atopic dermatitis, atopic asthma, rhinoconjunctivitis, allergic rhinitis, Omen's syndrome, systemic sclerosis, and graft-versus-host disease ); Th1 lymphocyte-related disorders (eg, rheumatoid arthritis, multiple sclerosis, psoriasis, Sjögren's syndrome, Hashimoto's thyroiditis, Grave's disease, primary biliary cirrhosis, Wegener's granulomatosis, and tuberculosis); and activated B lymphocyte-related disorders (eg, systemic lupus erythematosus, Goodpasture syndrome, rheumatoid arthritis, and type I diabetes).

Multidrug therapy for autoimmune diseases

Also disclosed are methods of treating an autoimmune disease, comprising administering to a patient in need thereof an effective amount of a ligand-drug conjugate and another known therapeutic agent for treating the autoimmune disease.

Compositions and methods of administration

The present invention provides pharmaceutical compositions comprising the ligand-drug conjugates described herein and a pharmaceutically acceptable carrier. Ligand-drug conjugates can be in any form that allows the compound to be administered to a patient to treat a disorder associated with the expression of an antigen to which the ligand moiety binds. For example, the conjugates may be in liquid or solid form. The preferred route of administration is parenteral. Parenteral administration includes subcutaneous injection, intravenous, intramuscular, intrasternal injection or infusion. In one aspect, the compositions are administered parenterally. In one aspect, the conjugates are administered intravenously. Administration can be carried out by any convenient route, for example, by infusion or bolus injection.

Pharmaceutical compositions can be formulated such that the compound is bioavailable when the composition is administered to a patient. The compositions may be in the form of one or more dosage units, where, for example, a tablet may be one dosage unit.

The substances used to prepare the pharmaceutical compositions may be non-toxic in the amounts used. It will be apparent to those of ordinary skill in the art that the optimal dosage of the active ingredient(s) in a pharmaceutical composition will depend on various factors. Such factors include, but are not limited to, the type of animal (eg, human), the specific form of the compound, the route of administration, and the composition used.

The composition may be, for example, in the form of a liquid.

Such a liquid may be useful for delivery by injection. The composition for administration by injection may also include one or more substances selected from a surfactant, a preservative, a wetting agent, a dispersing agent, a suspending agent, a buffer, a stabilizer and an isotonic agent.

Liquid compositions, whether in the form of solutions, suspensions or other similar form, may also include one or more of the following: sterile diluents such as water for injection, saline solution, preferably physiological saline solution,

- 94 046498 Ringer's solution, isotonic sodium chloride solution, fixed oils such as synthetic mono or diglycerides which can serve as a solvent or suspension medium, polyethylene glycols, glycerin, cyclodextrin, propylene glycol or other solvents; antibacterial agents such as benzyl alcohol or methylparaben; antioxidants such as ascorbic acid or sodium bisulfite; chelating agents such as ethylenediaminetetraacetic acid; buffers such as amino acids, acetates, citrates or phosphates; detergents such as nonionic surfactants, polyols; and agents for adjusting tonicity, such as sodium chloride or dextrose. The parenteral composition may be contained in an ampoule, a disposable syringe, or a multi-dose vial made of glass, plastic, or other material. Physiological saline solution is an example of an adjuvant. The injectable composition is preferably sterile.

The amount of conjugate that is effective for treating a particular disorder or condition will depend on the nature of the disorder or condition and can be determined by standard clinical techniques. In addition, in vitro or in vivo assays may optionally be used to determine optimal dosage ranges. The exact dosage for use in the compositions will also depend on the route of administration and the severity of the disease or disorder, and will be determined in accordance with the judgment of the attending physician and the specific circumstances of each patient.

The compositions include an effective amount of the compound so that a suitable dose can be obtained. Typically, this amount is at least about 0.01% of the compound, based on the weight of the composition.

For intravenous administration, the composition may include from about 0.01 to about 100 mg of ligand-drug conjugate per kg of animal body weight. In one aspect, the composition may include from about 1 to about 100 mg of ligand-drug conjugate per kg of animal body weight. In another aspect, the amount administered will range from about 0.1 to about 25 mg/kg body weight.

In general, the dose of conjugate that is administered to a patient is typically from about 0.01 mg/kg to about 100 mg/kg of the subject's body weight. In some embodiments, the dose administered to the patient is from about 0.01 mg/kg to about 15 mg/kg of the subject's body weight. In some embodiments, the dose administered to the patient is from about 0.1 mg/kg to about 15 mg/kg of the subject's body weight. In some embodiments, the dose administered to the patient is from about 0.1 mg/kg to about 20 mg/kg of the subject's body weight. In some embodiments, the dose administered is from about 0.1 mg/kg to about 5 mg/kg, or from about 0.1 mg/kg to about 10 mg/kg of the subject's body weight. In some embodiments, the dose administered is from about 1 mg/kg to about 15 mg/kg of the subject's body weight. In some embodiments, the dose administered is from about 1 mg/kg to about 10 mg/kg of the subject's body weight. In some embodiments, the dose administered is from about 0.1 to 4 mg/kg, even more preferably from 0.1 to 3.2 mg/kg, or even more preferably from 0.1 to 2.7 mg/kg body weight the subject's body throughout the treatment cycle.

The term carrier refers to the diluent, adjuvant or excipient with which the compound is administered. Such pharmaceutical carriers may be liquids such as water and oils, including oils of petroleum, animal, vegetable or synthetic origin, such as peanut oil, soybean oil, mineral oil, sesame oil. The carriers may be saline, gum acacia, gelatin, starch paste, talc, keratin, colloidal silica, urea. In addition, excipients, stabilizers, thickeners, lubricants and dyes can be used. In one embodiment, when administered to a patient, the compound or compositions and pharmaceutically acceptable carriers are sterile. Water is an example of a vehicle when the compounds are administered intravenously. Saline and aqueous dextrose solutions and glycerol solutions can also be used as liquid carriers, especially for injection solutions. Suitable pharmaceutical carriers also include excipients such as starch, glucose, lactose, sucrose, gelatin, malt, rice, flour, chalk, silica gel, sodium stearate, glycerol monostearate, talc, sodium chloride, skim milk powder, glycerin, propylene glycol, water, ethanol The compositions of the present invention, if desired, may also contain minor amounts of wetting or emulsifying agents or pH buffering agents.

In one embodiment, the conjugates are formulated according to routine procedures into a pharmaceutical composition adapted for intravenous administration to animals, in particular humans. Typically, carriers or drug-carrying media for intravenous administration are sterile isotonic aqueous buffer solutions. If necessary, the compositions may also include a solubilizing agent. Compositions for intravenous administration optionally include a local anesthetic, such as lignocaine, to relieve pain at the injection site. Typically, the ingredients are supplied either separately or mixed together in a unit dosage form, such as a dry lyophilized powder or an anhydrous concentrate in a sealed container such as an ampoule or sachet, indicating the amount of active ingredient. When the conjugate is intended for administration by infusion, it may be marketed, for example, with an infusion bottle containing sterile water or saline solution.

- 95 046498 solution with pharmaceutical purity. When the conjugate is administered by injection, a vial of sterile water for injection or saline may be provided so that the ingredients can be mixed before administration.

Pharmaceutical compositions are typically formulated as sterile, substantially isotonic compositions and in full compliance with all US Food and Drug Administration Good Manufacturing Practices (GMP) requirements.

Illustrative Methods

This application provides methods for preparing a drug-linker compound represented by the structure of formula (IV), (V) or (VI) described above, such method comprising step (a): contacting the intermediate linker compound represented by the structure of formula VII, VIII or IX described herein, with sufficient X'-D group to react with LP or AD' to thereby form an LP-X-D or AD-X-D group for each occurrence of LP and AD', where -X-D represents a releasable complex group attached to a drug group; and X'-D is a releasable complex group precursor attached to a drug group, wherein X' is capable of reacting with LP and/or AD'.

In some aspects, the linker drug thus prepared will have a structure represented by formula IVa, IVb, Va, Vb, Vc, VIa or VIb described herein, and the linker intermediate used in step (a) has a structure represented by formula VIIIa, VIIIb, VIIIc, VIIId, IXa or IXb described in this application.

The method may further comprise step (a'): deprotecting a suitably protected linker intermediate whose structure corresponds to formula VIIIa, VIIIb, VIIIc, VIIId, IXa or IXb, where t is O and where suitably protected AD' or LP has the structure where R ¹¹¹ is independently selected from hydrogen, p-hydroxybenzyl, methyl, isopropyl, isobutyl, _{sec -} butyl, -CH2OH, -CH(OH) _CH3 , -CH2CH2SCH3 _, -CH2CONH2, -CH2COOH, _-CH2 CH ₂ CONH ₂ , -CH ₂ CH ₂ COOH, (CH2)3NHC(=NH)NH2, -(CH2)3NH2, -(CH2)3NHCOCH3, -(CH2)3NHCHO, (CH2)4NHC(=NH)NH2, -(CH2)4NH2, -(CH2)4NHCOCH3, -(CH2)4NHCHO, (CH2)3NHCONH2, -(CH2)4NHCONH2, -CH2CH2CH(OH)CH2NH2, 2 pyridylmethyl-, 3-pyridylmethyl-, 4-pyridylmethyl *- sn ₂ ^ * - sn ₂ · with suitable protection when necessary, where RPR represents a suitable thiol protecting group and the wavy line indicates covalent attachment of a suitable AD' or LP protected group in the linker intermediate;

and in step (a) contacting the resulting unprotected product of formula VIIIa, VIIIb, VIIIc, VIIId, IXa or IXb from step (a') with an X'-D group, wherein X' consists of a maleimide group capable of reacting with a free thiol group AD ' or LP with the formation of a thio-substituted succinimide group.

Alternatively, the method may further comprise step a': deprotecting a precursor Linker Intermediate of formula VIIIa, VIIIb, VIIIe, VIIId, IXa or IXb having a structure where t is 1, and AD'-AD' or AD'- LP is suitably protected, where suitably protected AD'-AD' or AD'-LP has the structure _rPR wherein R ¹¹¹ is hydrogen or methyl and

- 96 046498

RPR represents a suitable thiol protecting group that is removed, and the wavy line indicates covalent attachment of a suitable AD' protected group in the linker intermediate; and in step (a) contacting the resulting unprotected product of formula VIIIa, VIIIb, VIIIe, VIIId, IXa or IXb from step (a') with an X'-D group, wherein X' consists of a maleimide-containing group capable of reacting with free thiol groups AD '-AD' or AD'-LP to form thio-substituted succinimide-containing groups.

This application provides methods for preparing a ligand-drug conjugate represented by the structure of formula I, II or III described above, such method comprising the following steps (a): contacting a ligand-linker compound represented by the structure For\G\rt P' mules X, XI or XII described above, with enough X'-D group to react with L or AD' to thus form an LP-X-D or AD-X-D group for each case of LP and AD', where X-D represents a releasable complex group attached to a drug group; and X'-D is a Release Complex Group precursor attached to a drug group, wherein X' is capable of reacting with LP and/or AD'.

An exemplary ligand-drug conjugate thus prepared has a structure represented by formula Ia, Ib, IIa, IIb, IIb, IIIa, or IIIb described above, and the ligand-linker compound has a structure represented by formula XIa, XIb, XIc, XId , XIIa or XIIb described above.

The method may further comprise step a': deprotecting an appropriately protected Ligand-Linker compound whose structure corresponds to formula X, XI, XII, XIa, XIb, XIc, XId, XIIa or XIIb described above, where t is 0 and where a suitably protected AD' or LP has the structure

where R is independently selected from hydrogen, p-hydroxybenzyl, methyl, isopropyl, isobutyl, sec-butyl, -CH2OH, -CH(OH)CH3, -CH2CH2SCH3, -CH2CONH2, -CH2COOH, -CH2CH2CONH2, -CH2CH2COOH, (CH2)3NHC(= NH)NH2, -(CH2)3NH2, -(СВДзХНСОНз, -(СН3^^СНО, (CH2)4NHC(=NH)NH2, -(CH2)4NH2, -(CH2)4NHCOCH3, -(C^^NHCHO, (CH^NHCON^, -(C^^NHCONH^ -CH2CH2CH(OH)CH2NH2, 2pyridylmethyl-, 3-pyridylmethyl-, 4-pyridylmethyl-, with suitable protection when necessary

where RPR represents a suitable thiol protecting group and the wavy line indicates the covalent attachment of a suitable AD' or LP protected group to a ligand-linker intermediate;

and in step (a) contacting the resulting unprotected product of formula X, XI, XII, XIa, XIb, XIc, XId, XIIa or XIIb from step (a') with an X'-D group, where X' consists of a maleimide group capable of react with the free thiol group of AD' or LP to form a thio-substituted succinimide group.

Alternatively, the method may further comprise step (a'): deprotection of a ligand linker compound whose structure corresponds to formula XIa, XIb, XIc, XId, XIIa or XIIb, where t is 1 and AD'-AD' or AD'-LP is suitably protected, wherein the suitably protected AD'-AD' group has the structure

where R ¹¹¹ represents hydrogen or methyl and RPR represents a suitable thiol protecting group that is removed, and the wavy line indicates the covalent attachment of a suitable AD' protected group in a ligand-linker intermediate; and at stage (a) contacting

- 97 046498 the resulting unprotected product of formula XIa, XIb, XIc, XId, Xlla or XIIb from step (a') with X'-D group, where X' consists of a maleimide-containing group capable of reacting with free thiol groups AD'-AD' or AD'-L ^P with the formation of thio-substituted succinimide-containing groups.

An exemplary maleimide group capable of reacting with the free thiol(s) from step (a') has the structure o

o where R ¹⁷ represents (CH ₂ ) ₅ C (=O)- and the wavy line indicates attachment in the X'-D group, or has the structure:

O

o where the wavy line indicates attachment at the X'-D group, and the amino group is optionally protected by an amino protecting group that is stable under deprotection conditions of R ^pr protected thiol groups.

Examples

General information.

All commercially available anhydrous solvents were used without further purification. PEG reagents were obtained from Quanta BioDesign (Powell, ON). Analytical thin layer chromatography was performed on silica gel 60 F254 coated aluminum plates (EMD Chemicals, Gibbstown, NJ). Radial chromatography was performed on a Chromatotron device (Harris Research, Palo Alto, CA).

Flash column chromatography was performed on a Biotage Isolera One purification system (Charlotte, NC). Analytical HPLC was performed on a Varian ProStar 210 solvent delivery system configured with a Varian ProStar 330 PDA detector. Samples were eluted through a Phenomenex Synergi 2.0 x 150 mm, 4 μm, 80 A C12 reverse phase column. The acidic mobile phase consisted of acetonitrile and water containing either 0.05% trifluoroacetic acid or 0.1% formic acid (indicated for each connections). Compounds were eluted with a linear gradient of acidic acetonitrile from 5% at 1 min after sample injection to 95% at 11 min, followed by isocratic elution with 95% acetonitrile at 15 min (flow rate=1.0 ml/min). LC-MS was performed on two different systems. LC-MS system 1 consisted of a ZMD Micromass mass spectrometer coupled to an HP Agilent 1100 HPLC unit equipped with a C12 Phenomenex Synergi 2.0 x 150 mm, 4 µm, 80 A reverse phase column. The acid eluent consisted of a linear gradient of 5% to 95% acetonitrile in 0.1% aqueous formic acid over 10 min, followed by an isocratic gradient of 95% acetonitrile over 5 min (flow rate=0.4 ml/min). LC-MS system 2 consisted of a Waters Xevo G2 Tof mass spectrometer coupled to a Waters 2695 Separations Module with a Waters 2996 Photodiode Array Detector; column, mobile phases, gradient, and flow rate were the same as for LC-MS system 1.

LC-MS data for antibody-drug conjugates were obtained on a Waters Xevo GS-S QTOF system coupled to a Waters Acquity H-Class HPLC system. Samples were chromatographed on a reverse phase analytical column (Agilent Technologies, PLRP-S, 300A, 2.1 mm ID x 50 mm, 8 µm) at 80°C and eluted with a linear gradient of 0.01% TFA in acetonitrile from 25 % to 65% in 0.05% aqueous TFA over 12.5 minutes, followed by isocratic elution with 65% 0.01% TFA in acetonitrile over 1.5 minutes at a flow rate of 1 ml/min. Mass spectrometry data for the light and heavy chains were acquired in ESI+ mode using the mass range of 500-4000 m/z and deconvolved using MaxEntl to determine the masses of the resulting conjugates.

Preparative HPLC was performed on a Varian ProStar 210 solvent delivery system configured with a Varian ProStar 330 PDA detector. Products were purified on a Phenomenex Synergi 10.0 x 250 mm, 4 μm, 80 A reverse phase C12 column, eluting with 0.1% formic acid in water (solvent A) and 0.1% formic acid in acetonitrile (solvent B). The purification method consisted of the following gradient of solvent A to solvent B: 90:10 from 0 to 5 min; 90:10 to 10:90

- 98 046498 from 5 minutes to 80 minutes; followed by isocratic 10:90 for 5 min. The flow rate was 4.6 ml/min and recorded at 254 nm. Preparative HPLC for compounds in Schemes 3 and 4 was performed using 0.1% trifluoroacetic acid in both mobile phases instead of 0.1% formic acid. NMR spectra data were obtained on a Varian Mercury 400 MHz spectrometer. Coupling constants (J) are given in hertz.

Example 1. Synthesis of a glucuronide-MMAE drug-linker compound containing a PEG group in a sequential orientation

Scheme 1.

(2S,3S,4S,5R,6S)-6-(2-(3-aminopropanamido)-4-((5S,8S,11S,12R)-11-((S)-sec-butyl)-12- (2-((S)-2((1 R,2R)-3 -(((1 S,2R)-1 -hydroxy-1-phenylpropan-2-yl)amino)-1 -methoxy-2-methyl -3-oxopropyl)pyrrolidin-1-yl)-2-oxoethyl)-5,8-diisopropyl-4,10-dimethyl-3,6,9-trioxo-2,13-dioxa-4,7,10-triazatetradecyl )phenoxy)-3,4,5-trihydroxytetrahydro-2H-pyran-2-carboxylic acid (3):

The synthesis of compound 2 has been described previously (US Patent Publication 2008/0241128), which document is incorporated herein by reference. To a flask containing glucuronide-MMAE intermediate 2 (40 mg, 26.8 μmol), 0.9 mL of methanol and 0.9 mL of tetrahydrofuran were added. The solution was then cooled in an ice bath and lithium hydroxide monohydrate (6.8 mg, 161 μmol) was added dropwise as a solution in 0.9 mL of water. The reaction mixture was then stirred on ice for 1.5 hours when LC/MS analysis indicated complete conversion to product. Glacial acetic acid (9.2 µl, 161 µmol) was then added and the reaction mixture was concentrated to dryness. Preparative HPLC yielded fully deprotected glucuronide-MMAE linker intermediate 3 (26 mg, 87%) as an oily residue. Analytical HPLC (0.1% formic acid): t _R 9.3 min. LC-MS system 1: t _R 11.10 min, m/z (ES+) found 1130.48 (M+H) ⁺ , m/z (ES ^- ) found 1128.63 (M-H) ^- .

(28,38,48,5K,68)-6-(4-((58,88,118,12K)-11-((8)-sec-butyl)-12-(2-((8)-2- ((1K,2K)-3-(((18.2K)-1-hydroxy1-phenylpropan-2-yl)amino)-1-methoxy-2-methyl-3-oxopropyl)pyrrolidin-1-yl)-2 -oxoethyl)-5,8-diisopropyl-4,10-dimethyl-3,6,9-trioxo-2,13-dioxa-4,7,10-triazatetradecyl)-2-(1-(2,5-dioxo -2,5-dihydro-1H-pyrrol-1-yl)-3,79-dioxo-7,10,13,16,19,22,25,28,31,34,37,40,43,46, 49,52,55,58,61,64,67,70,73,76-tetracosaoxa4,80-diazatrioctacontanamido)phenoxy)-3,4,5-trihydroxytetrahydro-2H-pyran-2-carboxylic acid (4):

To a flask containing unprotected glucuronide-MMAE intermediate 3 (26 mg, 23 µmol) dissolved in anhydrous DMF (0.94 ml), maleimido-PEG24-KIB ester (32 mg, 23 µmol) was added as a solution in dimethylacetamide (200 mg/ml). Diisopropylethylamine (20 μL, 115 μmol) was added and the reaction mixture was stirred under nitrogen at ambient temperature for 6 hours when LC/MS analysis indicated conversion to the desired product. The reaction mixture was purified by preparative HPLC to obtain linear maleimido-PEG24-glucuronide

- 99 046498

MMAE linker 4 (31 mg, 55%) as an oily residue. 1H NMR (CD ₃ OD) δ (ppm) 0.92 (m, 16H), 1.15 (m, 6H), 1.42 (m, 2H), 1.60 (m, 2H), 1.91 (m, 4H), 2.20 (m, 3H), 2.48 (m, 6H), 2.66 (m, 3H), 2.96 (m, 4H), 3.10 (s , 2H), 3.27 (s, 2H), 3.31 (s, 8H), 3.38 (m, 5H), 3.44 (m, 2H), 3.57 (m, 6H), 3 .62 (m, 79Н), 3.77 (m, 5Н), 3.87 (t, J=9, 6 Hz, 2Н), 4.05 (m, 1Н), 4.21 (m, 3Н) , 4.53 (m, 2H), 4.61 (m, 2H), 4.80 (m, 2H), 5.14 (m, 3H), 6.82 (s, 2H), 7.10 ( m, 2H), 7.21 (m, 2H), 7.35 (m, 2H), 7.39 (m, 2H), 7.74 (d, J=8.8 Hz, 1H), 7, 94 (m, 2H), 8.10 (m, 1H), 8.27 (m, 2H). Analytical HPLC (0.1% formic acid): tR 9.9 min. LC-MS system 1: t _R 11.94 min, m/z (ES+) found 1205.34 (M+2H) ²⁺ . LC-MS system 2: t _R 10.38 min, m/z (ES ⁺ ) found 2410.3225 (M+H) ⁺ .

Example 2. Synthesis of a glucuronide-MMAE drug-linker compound containing a PEG group in parallel orientation

Scheme 2.

^)-80-((((9H-fluoren-9-yl)methoxy)carbonyl)amino)-74-oxo2,5,8,11,14,17,20,23,26,29,32,35, 38,41,44,47,50,53,56,59,62,65,68,71-tetracosaoxa-75-azagenoctacontane-81-oic acid (6):

To a flask containing Na-Fmoc-lysine 5 (30 mg, 81.5 µmol), 1.6 ml of anhydrous dichloromethane was added, followed by the addition of methoxy-PEG-24-OSu (100 mg, 81.5 µmol). DIPEA (71 μL, 408 μmol) was then added and the reaction mixture was stirred under nitrogen at room temperature followed by TLC and LC/MS analysis. After 2 hours, LC/MS analysis showed conversion to product. The reaction solution was diluted in dichloromethane and immediately applied to a 1 mm chromatography plate for purification. The plate was eluted with dichloromethane with increasing amounts of methanol (0% to 15%) to give the desired product 6 (63 mg, 53%). TLC: Rf=0.17, 10% MeOH in CH2a2. 1H NMR (CDCl ₃ ) δ (ppm) 1.48 (m, 6H), 2.47 (m, 5H), 3.20 (m, 2H), 3.38 (s, 3H), 3 .63 (m, 86Н), 4.16 (m, 2Н), 4.36 (m, 1Н), 7.26 (m, 3Н), 7.35 (m, 2Н), 7.60 (m, 2H), 7.71 (m, 3H). Analytical HPLC (0.1% formic acid): t _R 10.8 min. LC-MS system 1: t _R 11.95 min, m/z (ES ⁺ ) found 1468.40 (M+H) ⁺ , m/z (ES ^- ) found 1466.36 (M-H) ^- .

^)-2,5-dioxopyrrolidin-1-yl 80-((((9H-fluoren-9-yl)methoxy)carbonyl)amino)-74-oxo2,5,8,11,14,17,20,23 ,26,29,32,35,38,41,44,47,50,53,56,59,62,65,68,71-tetracosaoxa-75-azagenoctacontane81-oate (7):

The flask was charged with Nc-Fmoc-lysine (PEG24)-OH 6 (63 mg, 43 μmol) and 0.43 ml of anhydrous tetrahydrofuran. N-hydroxysuccinimide (5.5 mg, 47 µmol) was added followed by diisopropylcarbodiimide (7.3 µl, 47 µmol). The reaction vessel was sealed and the mixture was stirred under nitrogen atmosphere overnight. After 18 hours, additional Nhydroxysuccinimide (5.5 mg, 47 µmol) and diisopropylcarbodiimide (7.3 µl, 47 µmol) were added and mixed

- 100 046498 cross-linking was continued for a further 4 hours when LC/MS analysis showed complete conversion to product. The crude product was diluted in dichloromethane and purified by radial chromatography on a 1 mm plate, eluting with dichloromethane with increasing amounts of methanol (0 to 10%) to give the desired activated ester 7 (36 mg). This substance was used further without additional characterization. TLC: Rf=0.43, 10% MeOH in CH2Cl2. Analytical HPLC (0.1% formic acid): t _R 11.4 min. LC-MS system 2: t _R 11.01 min, m/z (ES ⁺ ) found 1564.8379 (M+H) ⁺ .

(28,38,48,5I,68)-6-(2-((8)-80-((((9H-fluoren-9-yl)methoxy)carbonyl)amino)-74,81-dioxo2,5 ,8,11,14,17,20,23,26,29,32,35,38,41,44,47,50,53,56,59,62,65,68,71-tetracosaoxa-75,82 -diazaleneoctacontanamido)-4-((5S,8S,11S,12R)-11-((S)-sec-butyl)-12-(2-((S)-2-((1R,2R)-3- (((1S,2R)-1-hydroxy-1phenylpropan-2-yl)amino)-1-methoxy-2-methyl-3-oxopropyl)pyrrolidin-1-yl)-2-oxoethyl)-5,8-diisopropyl -4,10-dimethyl-3,6,9-trioxo-2,13-dioxa-4,7,10-triazatetradecyl)phenoxy)-3,4,5-trihydroxytetrahydro-2H-pyran-2-carboxylic acid (8 ):

Unprotected glucuronide-MMAE linker intermediate 3 (26 mg, 23 μmol) was dissolved in anhydrous dimethylformamide (0.58 ml) and added to the flask containing N,,-Fmoclysine(PEG)-OSu 7 (36 mg, 23 μmol). Diisopropylethylamine (20 μL, 115 μmol) was then added and the reaction mixture was then stirred under nitrogen at room temperature. After 4.5 h, LCMS analysis showed conversion to product. The product was purified by preparative HPLC to give Fmoc-Lys(PEG24)-glucuronide-MMAE intermediate 8 (30 mg, 50% of two steps) as an oily residue. Analytical HPLC (0.1% formic acid): t _R 11.4 min. LC-MS system 1: t _R 12.31 min, m/z (ES ⁺ ) found 1291.05 (M+2H) ²⁺ . LC-MS system 2: t _R 11.30 min, m/z (ES ⁺ ) found 2580.2515 (M+H) ⁺ .

(28,38,48,5I,68)-6-(2-((8)-80-amino-74,81-dioxo2,5,8,11,14,17,20,23,26,29, 32,35,38,41,44,47,50,53,56,59,62,65,68,71-tetracosaoxa-75,82diazanent-octacontanamido)-4-((5S,8S,11S,12R)-11- ((S)-sec-butyl)-12-(2-((S)-2-((1R,2R)-3-(((1S,2R)-1-hydroxy-1-phenylpropan-2-yl )amino)-1-methoxy-2-methyl-3-oxopropyl)pyrrolidin-1-yl)-2-oxoethyl)-5,8-diisopropyl-4,10-dimethyl-3,b,9-trioxo-2, 13-dioxa-4,7,10-triazatetradecyl)phenoxy)-3,4,5-trihydroxytetrahydro-2H-pyran-2-carboxylic acid (9):

Fmoc-Lys(PEG24)-glucuronide-MMAE intermediate 8 (30 mg, 12 μmol) was dissolved in 0.46 ml anhydrous dimethylformamide, followed by the addition of 0.12 ml piperidine. The reaction mixture was stirred under nitrogen atmosphere for 3 hours and then concentrated to dryness. The product was purified by preparative HPLC to give H-Lys(PEG24)-glucuronide-MMAE intermediate 9 (24 mg, 87%) as an oily residue. 1H NMR (CDCl ₃ ) δ (ppm) 0.92 (m, 14H), 1.14 (m, 6H), 1.42 (m, 5H), 1.79 (m, 8H), 2 .22 (m, 3H), 2.42 (t, J=6.4 Hz, 2H), 2.47 (m, 2H), 2.65 (m, 2H), 2.76 (m, 2H) , 2.95 (m, 3H), 3.10 (m, 3H), 3.31 (m, 8H), 3.35 (m, 6H), 3.54 (m, 5H), 3.63 ( s, 70H), 3.72 (t, J=6.0 Hz, 3H), 3.85 (m, 2H), 4.07 (m, 1H), 4.22 (m, 3H), 4, 52 (d, J=7.2 Hz, 1H), 4.61 (d, J=6.4 Hz, 1H), 4.71 (m, 2H), 5.11 (m, 3H), 7, 12 (m, 1H), 7.21 (m, 1H), 7.31 (m, 3H), 7.37 (m, 2H), 7.75 (d, J=8.8 Hz, 1H), 7.89 (d, J=8.8 Hz, 1H), 7.95 (d, J=8.8 Hz, 1H), 8.26 (m, 2H). Analytical HPLC (0.1% formic acid): t _R 8.9 min. LC-MS system 1: t _R 11.18 min, m/z (ES ⁺ ) found 1178.97 (M+2H) ²⁺ . LC-MS system 2: t _R 9.50 min, m/z (ES ⁺ ) found 2358.2341 (M+H) ⁺ .

^^^,5^)-6-(4-(^^,1^,^)-11-(^)-second-bugil)-12-(2-(^)-2-((^^) -3-(((^^)-1hydroxy-1-phenylpropan-2-yl)amino)-1-methoxy-2-methyl-3-oxopropyl)pyrrolidin-1-yl)-2-oxoethyl)5,8- diisopropyl-4,10-dimethyl-3,6,9-trioxo-2,13-dioxa-4,7,10-triazatetradecyl)-2-(^)-80-(6-(2,5dioxo-2,5 -dihydro-1H-pyrrol-1-yl)hexanamido)-74,81-dioxo2,5,8,11,14,17,20,23,26,29,32,35,38,41,44,47, 50,53,56,59,62,65,68,71-tetracosaoxa-75,82-diazapentaoctacontanamido)phenoxy)-3,4,5-trihydroxytetrahydro-2H-pyran-2-carboxylic acid (10): NHS maleimidocaproic ester acid (4.2 mg, 14 µmol) was dissolved in 0.6 ml anhydrous dimethylformamide and transferred to a flask containing H-Lys(PEG24)-glucuronide-MMAE intermediate 9 (24 mg, 10 µmol). Diisopropylethylamine (10 μL, 58 μmol) was then added and the reaction mixture was then stirred under nitrogen at room temperature overnight. The reaction mixture was purified directly by preparative HPLC to obtain MC-Lys(PEG24)-glucuronide-MMAE linker 10 (23 mg, 90%) as an oily residue. 1H NMR (CD ₃ OD) δ (ppm) 0.87 (m, 13H), 1.12 (t, J=7.6 Hz, 2H), 1.17 (d, J=6.8 Hz, 2H), 1.24 (m, 2H), 1.48 (m, 9H), 1.80 (m, 5H), 2.19 (m, 4H), 2.42 (t, J=6 .4 Hz, 2H), 2.48 (m, 2H), 2.64 (m, 2H), 2.96 (m, 3H), 3.10 (s, 1H), 3.12 (m, 2H ), 3.15 (s, 1H), 3.27 (s, 6H), 3.35 (m, 3H), 3.43 (m, 3H), 3.54 (m, 3H), 3.58 (m, 2H), 3.63 (m, 64H), 3.70 (m, 4H), 3.92 (m, 2H), 4.22 (m, 4H), 4.54 (m, 1H) , 4.61 (t, J=6.4 Hz, 1H), 4.83 (m, 1H), 5.13 (m, 3H), 6.80 (s, 2H), 7.10 (m, 1H), 7.20 (m, 2H), 7.29 (m, 2H), 7.38 (m, 2H), 7.74 (d, J=8.8 Hz, 1H), 7.90 ( m, 3H), 8.08 (s, 1H), 8.26 (m, 2H). Analytical HPLC (0.1% formic acid): t _R 10.6 min. LC-MS system 1: t _R 11.88 min, m/z (ES ⁺ ) found 1276.23 (M+2H) ²⁺ . LC-MS system 2: t _R 10.54 min, m/z (ES ⁺ ) found 2551.2871 (M+H) ⁺ .

- 101 046498

Example 3. Synthesis of mDPR (maleimido-diaminopropionic) glucuronide-MMAE drug-linker compound

Scheme 3 a.

In a 50 ml round bottom flask, H-DPR(boc)-OH and maleic anhydride were dissolved in 4 vol. acetic acid and the solution was stirred at room temperature for 3 hours. The reaction mixture was concentrated on a rotary evaporator to give an oil and the product was precipitated by adding -10 ml of dichloromethane. The precipitate was collected by vacuum filtration, washed with dichloromethane and dried overnight in a vacuum oven.

Maleyl-1)PI(boc)-O11 was suspended in toluene (3 ml) and triethylamine (224 µl) over molecular sieves in a 50 ml round bottom flask equipped with a condenser. DMA (-150 µl) was added to facilitate dissolution. The solution was heated to 125°C and refluxed for 4 hours, after which LC/MS analysis showed that the reaction was complete. The reaction mixture was concentrated to dryness on a rotary evaporator, redissolved in DMSO and purified by preparative HPLC. The product was isolated in the form of a white powder.

Scheme 3b.

Vos

Vos

55%

DMF, DIPEA room pace.

65% V

^)-2,5-dioxopyrrolidin-1-yl 3-((tert-butoxycarbonyl)amino)-2-(2,5-dioxo-2,5-dihydro-1Hpyrrol-1-yl)propanoate (12):

^)-Ma-maleimido-MB-Boc-diaminopropanoic acid 11 (Scheme 3a) (400 mg, 1.4 mmol) was dissolved in 7 ml of anhydrous dimethylformamide. N-hydroxysuccinimide (178 mg, 1.5 mmol) was added followed by 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide (298 mg, 1.5 mmol). The reaction mixture was stirred at room temperature under a nitrogen atmosphere for 3 hours. Aqueous treatment was carried out by diluting in 120 ml of water; the aqueous layer was then extracted three times at

- 102 046498 help with 60 ml ethyl acetate. The combined organic layer was then washed with brine, dried over sodium sulfate and concentrated to dryness. The product was purified by flash column chromatography, eluting with hexane:ethyl acetate (50:50 to 0:100), to give NHS ester

^)-M _a -maleimido-MB-Boc-diaminopropanoic acid [MDpr(Boc)-OSu] 12 (297 mg, 55%). LC-MS system 1: t _R 12.23 min, m/z (ES ⁺ ) found 282.0599 (M+H-Boc groups) ⁺ . LC-MS system 2: t _R 11.30 min, m/z (ES ⁺ ) found 2580.2515 (M+H) ⁺ .

(2S,3S,4S,5R,6S)-6-(2-(3-((s)-3-((tert-butoxycarbonyl)amino)-2-(2,5-dioxo-2,5-dihydro -1Hpyrrol-1-yl)propanamido)propanamido)-4-((5S,8S,11S,12R)-11-((s)-sec-butyl)-12-(2-((S)-2-( (1R,2R)-3(((1S,2R)-1-hydroxy-1-phenylpropan-2-yl)amino)-1-methoxy-2-methyl-3-oxopropyl)pyrrolidin-1-yl)-2oxoethyl )-5,8-diisopropyl-4,10-dimethyl-3,6,9-trioxo-2,13-dioxa-4,7,10-triazatetradecyl)phenoxy)3,4,5-trihydroxytetrahydro-2H-pyran- 2-carboxylic acid (13):

MDpr(Boc)-OSu 12 (33 mg, 86 μmol) was dissolved in 1.1 ml anhydrous dimethylformamide and added to the flask containing unprotected glucuronide-MMAE linker intermediate 3 (49 mg, 43 μmol). Diisopropylethylamine (37 μL, 220 μmol) was then added and the reaction mixture was then stirred under nitrogen at room temperature for 30 min. The reaction was quenched with 37 μl of glacial acetic acid and the mixture was purified by preparative HPLC to give MDpr(Boc)-glucuronide-MMAE intermediate 13 (39 mg, 65%). LC-MS system 2: t _R 11.09 min, m/z (ES ⁺ ) found 1396.7321 (M+H) ⁺ .

^^^^^)-6-(2-(3-(^)-3-amino-2-(2,5-dioxo-2,5-dihydro-1H-pyrrol-1-yl)propanamido)propanamido) -4-((5S,8S,11S,12R)-11-((S)-sec-butyl)-12-(2-((S)-2-((lR,2R)-3-((( 1S,2R)-1-hydroxy-1-phenylpropan-2-yl)amino)-1-methoxy-2-methyl-3-oxopropyl)pyrrolidin-1-yl)-2-oxoethyl)-5,8-diisopropyl4, 10-dimethyl-3,6,9-trioxo-2,13-dioxa-4,7,10-triazatetradecyl)phenoxy)-3,4,5-trihydroxytetrahydro2H-pyran-2-carboxylic acid (14):

The flask containing MDpr(Boc)-glucuronide-MMAE intermediate 13 (18 mg, 13 μmol) was cooled to 0° C. in an ice bath under nitrogen. A solution of 10% trifluoroacetic acid in dichloromethane (1.3 ml) was added dropwise. The reaction mixture was then stirred at 0°C for 2 hours, when LC-MS analysis showed complete removal of the Boc protecting group. The reaction mixture was then concentrated to give a crude residue and purified by preparative HPLC to give MDpr-glucuronide-MMAE linker 14 (15 mg, 92%). LC-MS system 2: t _R 9.13 min, m/z (ES ⁺ ) found 1296.6697 (M+H) ⁺ .

Example 4. Synthesis of a mDPR (maleimido-diaminopropionic) glucuronide-MMAE drug-linker compound containing a PEG group in parallel orientation

Scheme 4.

^^^^^)-6-(2-(^)-80-(^)-3-((tert-butoxycarbonyl)amino)-2-(2,5-dioxo-2,5-dihydro-1Hpyrrol- 1-yl)propanamido)-74,81-dioxo-2,5,8,11,14,17,20,23,26,29,32,35,38,41,44,47,50,53,56 ,59,62,65,6 8,71-tetracosaoxa-75,82-diazapentaoctacontanamido)-4-((5S,8S,11S,12R)-11-((S)-sec-butyl)-12-(2 -((S)2-((lR,2R)-3-(((1S,2R)-1-hydroxy-1-phenylpropan-2-yl)amino)-1-methoxy-2-methyl-3-oxopropyl )pyrrolidin-1-yl)-2-oxoethyl)-5,8-diisopropyl-4,10-dimethyl-3,6,9-trioxo-2,13-dioxa-4,7,10-triazatetradecyl)phenoxy)- 3,4,5-trihydroxytetrahydro-2H-pyran-2-carboxylic acid (15): MDpr(Boc)-OSu 12 (33 mg, 86 μmol) was dissolved in 0.66 ml of anhydrous dimethylformamide and added to the flask containing H- Lys(PEG24)-glucuronide-MMAE linker intermediate 9 (135 mg, 57 µmol).

- 103 046498

Diisopropylethylamine (50 µl, 290 µmol) was then added and the reaction mixture was then stirred under nitrogen at room temperature for 2.5 hours. The reaction was quenched with 50 µl glacial acetic acid and purified by preparative HPLC to give MDpr(Boc)-Lys( PEG24) glucuronide-MMAE intermediate 15 (86 mg, 58%). LC-MS system 2: t _R 11.71 min, m/z (ES+) found 2624.2004 (M+H) ⁺ .

(2S,3 S,4S,5R,6S)-6-(2-((S)-80-((S)-3 -amino-2-(2,5-dioxo-2,5-dihydro-1 N-pyrrol-1-yl)propanamido)-74,81-dioxo-2,5,8,11,14,17,20,23,26,29,32,35,38,41,44,47,50 ,53,56,59,62,65,68,71-tetracosaoxa-75,82diazapentaoctacontanamido)-4-((5 S,8S,11S,12R)-11 -(^)-sec-butyl)-12-( 2-((S)-2-(( 1 R,2R)-3 -(((1 S,2R)-1 hydroxy-1-phenylpropan-2-yl)amino)-1 -methoxy-2-methyl- 3-oxopropyl)pyrrolidin-1-yl)-2-oxoethyl)5,8-diisopropyl-4,10-dimethyl-3,6,9-trioxo-2,13-dioxa-4,7,10-triazatetradecyl)phenoxy )-3,4,5-trihydroxytetrahydro-2H-pyran-2-carboxylic acid (16): The flask containing MDpr(Boc)-Lys(PEG24)glucuronide-MMAE intermediate 15 (86 mg, 33 µmol) was cooled to 0°C in an ice bath under nitrogen atmosphere. A solution of 10% trifluoroacetic acid in dichloromethane (3.3 ml) was added dropwise. The reaction mixture was then stirred at 0°C for 2 h when LC-MS indicated complete removal of the Boc protecting group. The reaction mixture was then concentrated to give a crude residue and purified by preparative HPLC to give MDpr-Lys(PEG24)-glucuronide-MMAE linker 16 (38 mg, 46%). LC-MS system 2: t _R 10.54 min, m/z (ES ⁺ ) found 2524.2256 (M+H) ⁺ .

Example 5. Synthesis of an mDPR (maleimido-diaminopropionic) glucuronide-MMAE drug-linker compound containing PEG12, PEG8 or PEG4-(PEG4) ₃ group in parallel orientation

Scheme 5.

(2S,3S,4S,5R,6S)-6-(2-((S)-44-((R)-3-amino-2-(2,5-dioxo-2,5-dihydro-1H- pyrrol-1-yl)propanamido)-3 8,45-dioxo-2,5,8,11,14,17,20,23,26,29,32,35-dodecaoxa-39,46-diazanonatetracontanamido)-4 ((5S,8S,11S,12R)-11 -((8)-sec-butyl)-12-(2-((S)-2-(( 1 R,2R)-3 -(((1 S ,2R)-1 -hydroxy-1-phenylpropan-2yl)amino)-1-methoxy-2-methyl-3-oxopropyl)pyrrolidin-1-yl)-2-oxoethyl)-5,8-diisopropyl-4,10dimethyl -3,6,9-trioxo-2,13-dioxa-4,7,10-triazatetradecyl)phenoxy)-3,4,5-trihydroxytetrahydro-2Hpyran-2-carboxylic acid (17):

MDpr-Lys(PEG12)-glucuronide-MMAE linker 17 was prepared in a manner similar to the method for preparing compound 16 described in Schemes 2 and 4. LC-MS system 2: t _R 9.88 min, m/z (ES ⁺ ) found 1996 .1001 (M+N) ⁺ .

(2S,3S,4S,5R,6S)-6-(2-((S)-32-((R)-3-amino-2-(2,5-dioxo-2,5-dihydro-1H- pyrrol-1-yl)propanamido)-26,33-dioxo-2,5,8,11,14,17,20,23-octaoxa-27,34-diazaheptatriacontanamido)-4((5S,8S,11S,12R )-11 -(^)-sec-butyl)-12-(2-((S)-2-(( 1 R,2R)-3 -(((1 S,2R)-1 -hydroxy-1 - phenylpropan-2yl)amino)-1-methoxy-2-methyl-3-oxopropyl)pyrrolidin-1-yl)-2-oxoethyl)-5,8-diisopropyl-4,10dimethyl-3,6,9-trioxo-2 ,13-dioxa-4,7,10-triazatetradecyl)phenoxy)-3,4,5-trihydroxytetrahydro-2Hpyran-2-carboxylic acid (18):

- 104 046498

MDpr-Lys(PEG8)-glucuronide-MMAE linker 17 was prepared in a manner similar to the method for preparing compound 16 described in Schemes 2 and 4. LC-MS system 2: t _R 10.50 min, m/z (ES ⁺ ) found 1818 .8678 (M+H) ⁺ .

(^^^.5K.6^)-6-(2-((b)-48-((K)-3-amino-2-(2.5-dioxo-2.5-dihydro-1H-pyrrol-1-yl )propanamido)-15,22,38,42,49-pentaoxo-20,20-bis(15-oxo-2,5,8,11,18-pentaoxa-14-azanonedecane-19-yl)2.5.8.11. 18.25.28.31.34-nonaoxa-14.21.37.43.50-pentaazatripentacontanamido)-4-((^^.1^.12K)-11(^)-sec-butyl)-12-(2-(^)-2 -((1K,2K)-3-(((^,2K)-1-hydroxy-1-phenylpropan-2-yl)amino)-1-methoxy-2methyl-3-oxopropyl)pyrrolidin-1-yl)- 2-oxoethyl)-5,8-diisopropyl-4,10-dimethyl-3,6,9-trioxo-2,13dioxa-4,7,10-triazatetradecyl)phenoxy)-3,4,5-trihydroxytetrahydro-2H- pyran-2-carboxylic acid (19):

MDpr-Lys(PEG4[PEG4]3)-glucuronide-MMAE linker 19 was prepared in a manner similar to the method for preparing compound 16 described in Schemes 2 and 4. LC-MS system 2: t _R 9.92 min, m/z (ES ⁺ ) found 2674.3813 (M+H) ⁺ .

^^^,5K^)-6-(2-(^)-20-((K)-3-amino-2-(2,5-dioxo-2,5-dihydro-III-pyrrol-1yl)propanamido )-14,21-dioxo-2,5,8,11-tetraoxa-15,22-diazapentacosanamido)-4-(^^,1^,12K)-11(^)-sec-butyl)-12-( 2-(^)-2-((1K,2K)-3-(((^,2K)-1-hydroxy-1-phenylpropan-2-yl)amino)-1-methoxy-2methyl-3-oxopropyl) pyrrolidin-1-yl)-2-oxoethyl)-5,8-diisopropyl-4,10-dimethyl-3,6,9-trioxo-2,13dioxa-4,7,10-triazatetradecyl)phenoxy)-3,4 ,5-Trihydroxytetrahydro-2H-pyran-2-carboxylic acid (42): MDpr-Lys(PEG4)-glucuronide-MMAE linker 42 was prepared in a manner similar to the preparation of compound 16 described in Schemes 2 and 4. LC-MS system 2 : t _R 10.18 min, m/z (ES ⁺ ) found 1642.8586 (M+H) ⁺ .

^^^^^)-6-(2-(^)-14-(^)-3-amino-2-(2,5-dioxo-2,5-dihydro-1H-pyrrol-1-yl)propanamido )-8,15-dioxo-2,5-dioxa-9,16-diazanonadecanamido)-4-((5S,8S,11S,12R)-11-((S)-sec-butyl)-12-(2 ((S)-2-((1R,2R)-3 -(((1S,2R)-1-hydroxy-1-phenylpropan-2-yl)amino)-1-methoxy-2-methyl-3-oxopropyl )pyrolidin-1-yl)-2-oxoethyl)-5,8-diisopropyl-4,10-dimethyl-3,6,9-trioxo-2,13-dioxa-4,7,10-triazatetradecyl)phenoxy)- 3,4,5-trihydroxytetrahydro-2H-pyran-2-carboxylic acid (43): MDpr-Lys(TOr2)glucuronide-MMAE linker 43 was prepared in a manner similar to the preparation of compound 16 described in Schemes 2 and 4. LC-MS system 2: t _R 10.10 min, m/z (ES ⁺ ) found 1554.8093 (M+H) ⁺ .

(^^^.5K.6^)-6-(2-(3-((b)-6-acetamido-2-((K)-3-amino-2-(2.5-dioxo-2.5-dihydro- 1H-pyrrol-1yl)propanamido)hexanamido)propanamido)-4-((5S,8S,11S,12R)-11-((S)-sec-butyl)-12-(2-((S)-2( (1R,2R)-3 -(((1 S,2R)-1-hydroxy-1-phenylpropan-2-yl)amino)-1-methoxy-2-methyl-3-oxopropyl)pyrrolidin-1-yl) -2-oxoethyl)-5,8-diisopropyl-4,10-dimethyl-3,6,9-trioxo-2,13-dioxa-4,7,10-triazatetradecyl)phenoxy)-3,4,5-trihydroxytetrahydro -2H-pyran-2-carboxylic acid (44):

MDpr-Lys(Ac)-glucuronide-MMAE linker 44 was prepared in a manner similar to the method for preparing compound 16 described in Schemes 2 and 4. LC-MS system 2: t _R 10.38 min, m/z (ES ⁺ ) found 1466 .8109 (M+H) ⁺ .

- 105 046498

Example 6. Synthesis of mDPR (maleimido-diaminopropionic) valine-citrulline-MMAE drug-linker compound containing a PEG group in parallel orientation

Scheme 6.

/ 23 R = Sun

10% TFA/DCM (

40% 24 R = n /21 R = Fmoc piperidine ί

93% 22 R = n

4-((808,838,868)-80-((((9H-fluoren-9-yl)methoxy)carbonyl)amino)-83-isopropyl-74,81,84-trioxo86-(3-ureidopropyl)-2,5, 8,11,14,17,20,23,26,29,32,35,38,41,44,47,50,53,56,59,62,65,68,71-tetracosaoxa75,82,85- triazaheptaoctacontanamido)benzyl ((8)-1-(((8)-1-(((3K,48,58)-1-((8)-2-((1K,2K)-3-(((18 ,2K)-1-hydroxy-1-phenylpropan-2-yl)amino)-1methoxy-2-methyl-3-oxopropyl)pyrrolidin-1-yl)-3-methoxy-5-methyl-1-oxoheptane-4- yl)(methyl)amino)3-methyl-1-oxobutan-2-yl)amino)-3-methyl-1-oxobutan-2-yl)(methyl)carbamate (21):

ValCit-PAB-MMAE linker (synthesized as described in US Patent No. 7659241) intermediate 20 (16 mg, 14 μmol) was dissolved in anhydrous dimethylformamide (0.28 ml) and added to the flask containing \a-1 ^; tos-lysine(1 G)G)-(.')8i 7 (25 mg, 17 µmol). Diisopropylethylamine (12 μL, 70 μmol) was then added and the reaction mixture was then stirred under nitrogen at room temperature. After 6 h, LC-MS analysis showed conversion to product. The product was purified by preparative HPLC to give Fmoc-Lys(PEG24)-ValCit-PAB-MMAE intermediate 21 (15 mg, 42%) as an oily residue. Analytical HPLC (0.1% formic acid): LCMS system 2: t _R 11.67 min, m/z (E8+) found 2573.2493 (M+H)+.

4-((808,838,868)-80-amino-83-isopropyl-74,81,84-trioxo-86-(3-ureidopropyl)2,5,8,11,14,17,20,23,26,29, 32,35,38,41,44,47,50,53,56,59,62,65,68,71-tetracosaoxa-75,82,85-triazaheptaoctacontanamido)benzyl ((8)-1-(((8 )-1-(((3K,48.58)-1-((8)-2-((1K,2K)-3-(((18.2K)-1-hydroxy-1-phenylpropane-2- yl)amino)-1methoxy-2-methyl-3-oxopropyl)pyrrolidin-1-yl)-3-methoxy-5-methyl-1-oxoheptan-4-yl)(methyl)amino)3-methyl-1-oxobutane -2-yl)amino)-3-methyl-1-oxobutan-2-yl)(methyl)carbamate (22):

Fmoc-Lys(PEG24)-ValCit-PAB-MMAE intermediate 21 (15 mg, 6 μmol) was dissolved in 0.16 ml anhydrous dimethylformamide, followed by the addition of 0.04 ml piperidine. The reaction mixture was stirred under nitrogen atmosphere for 1.5 hours and then concentrated to dryness. The product was purified by preparative HPLC to give H-Lys(PEG24)-ValCit-PAB-MMAE intermediate 22 (13 mg, 93%) as an oily residue. LC-MS system 2: t _R 9.72 min, m/z (E8 ⁺ ) found 2351.1787 (M+H) ⁺ .

4-((808,838,868)-80-((8)-3-((tert-butoxycarbonyl)amino)-2-(2,5-dioxo-2,5-dihydro-1H-pyrrol-1yl)propanamido)-83 -isopropyl-74,81,84-trioxo-86-(3-ureidopropyl)2,5,8,11,14,17,20,23,26,29,32,35,38,41,44,47, 50,53,56,59,62,65,68,71-tetracosaoxa-75,82,85-triazaheptaoc

- 106 046498 tacontanamido)benzyl ((S)-1 -(((S)-1 -(((3R,4S,5S)-1 -((S)-2-(( 1 R,2R)-3- ((( 1 S,2R)-1 -hydroxy-1 -phenylpropan-2-yl)amino)-1 methoxy-2-methyl-3-oxopropyl)pyrrolidin-1 -yl)-3-methoxy-5-methyl- 1-oxoheptan-4-yl)(methyl)amino)3-methyl-1-oxobutan-2-yl)amino)-3-methyl-1-oxobutan-2-yl) (methyl)carbamate (23):

MDpr(Boc)-OSu 12 (4 mg, 11 μmol) was dissolved in 0.12 ml anhydrous dimethylformamide and added to a flask containing H-Lys(PEG24)-ValCit-PAB-MMAE linker intermediate 22 (13 mg, 5. 5 µmol). Diisopropylethylamine (5 µl, 28 µmol) was then added and the reaction mixture was then stirred under nitrogen at room temperature for 1 h. The reaction was quenched with 5 µl glacial acetic acid and purified by preparative HPLC to give MDpr(Boc)Lys(PEG24)- ValCit-PAB-MMAE intermediate 23 (10 mg, 69%). LC-MS system 2: t _R 11.25 min, m/z (ES ⁺ ) found 2617.3203 (M+H) ⁺ .

4-((80S,83S,86S)-80-((S)-3-amino-2-(2,5-dioxo-2,5-dihydro-1H-pyrrol-1-yl)propanamido)-83isopropyl- 74,81,84-trioxo-86-(3-ureidopropyl)2,5,8,11,14,17,20,23,26,29,32,35,38,41,44,47,50,53 ,56,59,62,65,68,71-tetracosaoxa-75,82,85-triazaheptaoctacontanamido)benzyl ((S)-1 -(((S)-1 -(((3R,4S,5S)-1 -((S)-2-(( 1 R,2R)-3-((( 1 S,2R)-1 -hydroxy-1-phenylpropan-2-yl)amino)-1 methoxy-2-methyl-3 -oxopropyl)pyrrolidin-1 -yl)-3-methoxy-5-methyl-1 -oxoheptan-4-yl)(methyl)amino)3 -methyl-1 -oxobutan-2-yl)amino)-3-methyl- 1-oxobutan-2-yl)(methyl)carbamate (24):

The flask containing MDpr(Boc)-Lys(PEG24)-ValCit-PAB-MMAE intermediate 23 (10 mg, 4 μmol) was cooled to 0°C in an ice bath under nitrogen. A solution of 10% trifluoroacetic acid in dichloromethane (0.4 ml) was added dropwise. The reaction mixture was then stirred at 0°C for 3 hours. The reaction mixture was then concentrated to obtain a crude residue and purified by preparative HPLC to obtain MDpr-Lys(πEG24)-ValCit-PAB-MMAE linker 24 (4 mg, 40%). LCMS system 2: t _R 9.81 min, m/z (ES ⁺ ) found 2517.2930 (M+H) ⁺ .

Example 7: ADCs incorporating PEG in a parallel orientation exhibit in vitro activity similar to their non-PEGylated counterparts or ADCs incorporating PEG in a serial orientation

Cells that were cultured in log growth phase were seeded in 96-well plates containing 150 μl of RPMI 1640 supplemented with 20% FBS for 24 h. Serial dilutions of ADC in cell culture medium were prepared at 4x working concentrations; 50 μl of each dilution was added to 96-well plates. After addition of ADC, cells were incubated with test compounds for 4 days at 37°C. After 96 h, growth inhibition was assessed using Cell Titer Glo (Promega, Madison, WI) and luminescence was measured on a plate reader. The IC ₅₀ value, determined in triplicate, is here defined as the concentration that results in a 50% reduction in cell growth compared to untreated controls.

Compounds 1, 4 and 10 were conjugated via their interchain thiols to the chimeric cAC10 antibody described in US Patent No. 7,090,843, with an average drug loading of 8 drug groups per antibody. Compounds 4 and 10 are described above. Compound 1 is as follows:

Me OMEO CH ₃ O O

In vitro cytotoxic activity of the resulting ADCs was measured against CD30 + and CD30 ^- cell lines. Neither the addition of PEG nor its configuration had a significant effect on in vitro activity; Only a minor difference in ADC activity was observed, and in the two cell lines (L540cy and Karpas-299) the activity was essentially the same (Table 1).

- 107 046498

Table 1. In vitro cytotoxic activity of aHTu-CD30 ADCs; values represent IC ₅₀ values in ng/ml

CD30+ cell lines CD30- ADC drug/AB Karpas 299 L540cy L428 WSU-NHL SASYU-1 8 2.5 4.4 9 no effect SASYU-4 8 1.5 4.4 34 no effect SASYU-10 8 1.7 6, 6 13 no effect

Example 8: ADCs Including PEG in Parallel Orientation Show Favorable Pharmacokinetics Compared to ADCs Including PEG in Serial Orientation

Radiolabeling of Antibody and ADC - Pharmacokinetic (PK) experiments were performed using radiolabeled antibody or ADC. For RK experiments, test substances were radiolabeled using the following procedure. To a solution of the antibody or ADC in PBS supplemented with an additional 50 mM potassium phosphate (pH 8.0) and 50 mM sodium chloride, 55 μCi of N-succinimidylpropionate, [propionate- ^2,3-3H ]- (Moravek Biochemicals, Cat. No.: MT 919, 80 Ci/mmol, 1 mCi/ml, 9:1 hexane:ethyl acetate solution) per mg of antibody or ADC. The resulting mixture was vortexed and left at room temperature for 2 hours. The mixture was centrifuged at 4000xg for 5 minutes and the lower aqueous layer was removed and separated using Amicon Ultra-15 centrifugal filters (Millipore, Cat. No.: UFC903024, 30 kDa MWCO). Unbound radioactivity was removed using 4 cycles of dilution and centrifuged at 4000xg. The resulting products were filtered through sterile 0.22 μm Ultrafree-MC centrifuge filters (Millipore, Cat. No.: UFC30GV0S) and the final concentration of antibody or ADC was measured spectrophotometrically. The specific radioactivity (μCi/mg) of each product was determined by liquid scintillation counting.

Pharmacokinetic Experiments - The pharmacokinetic properties of the unconjugated antibody or ADC have been studied in several rodent models. In each experiment, 1-3 mg of radiolabeled antibody or ADC per kg of animal body weight was administered by tail vein injection. Each test agent was administered once to several animals in parallel. Blood was drawn into K ₂ EDTA tubes through the saphenous vein or through cardiac puncture to collect terminal blood at different time points. Plasma was separated by centrifugation for 10 min at 10,000xg. 10-20 μl of plasma sample from each time point was added to 4 ml of Ecoscint-A liquid scintillation cocktail (National Diagnostics) and total radioactivity was measured by liquid scintillation counting. The resulting dpm values were converted to µCi and the specific radioactivity of the radiolabeled test substances was used to calculate the concentration of antibody or ADC remaining in the plasma at each time point. Pharmacokinetic parameters (clearance and AUC) were determined from the obtained plasma concentration data. The pharmacokinetic parameters assessed were calculated using non-compartmental analysis in Phoenix WinNonlin v6.3 (Pharsight, Mountain View, CA) using the intravenous bolus dose option.

Compounds 1, 4 and 10 were conjugated via their interchain thiols to the chimeric cAC10 antibody described in US Pat. No. 7,090,843, which is incorporated herein by reference, at an average drug loading of 8 drug groups per antibody. As expected, ADCs made with 8 copies of non-PEGylated drug linker 1 exhibited very rapid clearance and low exposure compared to the unconjugated antibody (Figure 7). Surprisingly, PEGylated drug-linker 4 containing PEG in a sequential configuration produced ADCs with even faster clearance and lower exposure than the non-PEGylated format. This result was unexpected given the number of prior art examples of ADCs produced in accordance with this design. In contrast, ADCs prepared with drug linker 10 containing PEG in a parallel configuration produced ADCs with significantly slower clearance and greater exposure compared to the non-PEGylated format (see Figure 7 and Table 2).

- 108 046498

table 2

Conjugate Ligand Drug Clearance (ml/day/kg) AUCo-inf (day* µg/ml) SASYU 8, 6 604.1 SASYU-1 48, 6 67.0 SASYU-4 57.8 52.0 SASYU-10 14.2 229, 7

Alternatively, an ELISA-based total antibody (Tab) assay can be used to obtain pharmacokinetic data. 100 μl of a solution of anti-human IgG kappa antibody (0.5 mg/ml, Antibody Solutions, Mountain View CA) in 0.05 M carbonate-bicarbonate buffer (pH 9.6, Sigma Aldrich, St. Louis, MO) was added to each well of a 96-well polystyrene plate coated with MaxiSorp™ (Sigma Aldrich, St. Louis, MO). The plates were incubated at 4°C overnight. After incubation, the plate was washed 3 times with PBS containing 0.05% Tween-20 (PBS-T). The wells were then blocked with PBS-T containing 1% bovine serum albumin at room temperature for at least 1 hour. After blocking, the plate was washed 3 times with PBS-T. Concentrated stock solutions of antibody or ADC standards (40 x concentration) were prepared in rat or mouse plasma to construct a standard curve. Plasma samples and standards were then diluted 1:40 in PBS-T. Diluted samples and standards (100 μl) were added to the wells of the ELISA plate and incubated at room temperature for 1 hour. After incubation, the samples were removed and the plate was washed 3 times with PBS-T. A solution of Peroxidase-AffiniPure F(ab')2 Fragment Goat Anti-Human IgG, Fcy Fragment Specific (Jackson ImmunoResearch Laboratories, Inc., West Grove, PA) was diluted 1:30,000 in PBS-T and 100 μl was added to each well. The plate was incubated at room temperature for 1 hour. After incubation, samples were removed and the plate was washed 3 times with PBS-T. SureBlueTMB Microwell Peroxidase Substrate solution (KPL, Inc. Gaithersburg, MD) was added to each well (100 μl). The plate was incubated at room temperature for 11-12 min and reactions were quenched using 100 μl of 1 N HCl solution. The plates were read at 450 nm on a Molecular Devices Spectromax plate reader.

Example 9.

ADCs incorporating PEG in a parallel orientation have superior in vivo activity compared to ADCs incorporating PEG in a serial orientation or ADCs containing no PEG group in an in vivo xenograft model. All experiments were performed in accordance with Institutional Animal Care and Use Committee requirements in a facility fully accredited by the Association for the Assessment and Accreditation of Laboratory Animal Care. Efficacy experiments were performed in xenograft models of Karpas 299 anaplastic large cell lymphoma, L540cy Hodgkin lymphoma, Ramos Burkitt lymphoma and MCF-7 breast cancer. Cell suspensions or tumor fragments were implanted subcutaneously into immunocompromised mice. Mice bearing MCF-7 tumors were simultaneously administered a slow-release 17B-estradiol tablet implanted subcutaneously. Mice were randomized to study groups when the average tumor volume reached approximately 100 mm ³ . ADC or controls were administered intraperitoneally once. Tumor volume as a function of time was determined using the formula (LxW ² )/2. Animals were sacrificed when the tumor volume reached 1000 mm ³ . For mice showing prolonged regression, the test was terminated at approximately day 100 postimplantation.

Initial studies were performed at 2 mg/kg (single dose) of each ADC for the L540cy model (FIG. 1) and at 0.6 mg/kg (single dose) for the Karpas-299 model (FIG. 2). Graphs showing changes in tumor volume over time are shown in Fig. 1 and 2. All linker drugs were conjugated through their interchain thiols to the chimeric cAC10 antibody described in US Pat. No. 7,090,843, which is incorporated herein by reference, with an average drug loading of 8 drug groups per antibody. In both models, ADCs produced with 1 ΟΑΟΟ^^ΡΑΒ^Ι^), non-PEGylated) and 10 (PEGylated construct in Scheme 2), cured all animals (5/5) in their dosing groups, whereas ADCs prepared with 4 , did not have any therapeutic effect, but only caused a slight slowdown in tumor growth. The reduced activity of cAC10-4 is consistent with its greatly reduced exposure observed in the pharmacokinetic study, as shown in FIG. 7. It was hypothesized that pharmacokinetically controlled differences in activity could also be observed between sAC10-1 and sAC10-10, but that lower doses would be required. Accordingly, the tests were repeated with both models at dose levels of ^1/2 and 1/2 of the doses used in the initial tests. For L54 0cy, a dose of 1 mg/kg provided complete cure (6/6) for cAC10-10 but only 2/6 cure for cAC10-1 (Fig. 3). At the 0.5 mg/kg dose, no cure was observed in any group; however, sAS10-10 provided longer

- 109 046498 slower tumor growth than cAC10-1 (Fig. 3). At both dose levels, L540cy the antitumor activity of compound cAC10-10 was greater than that of cAC10-1, consistent with their respective pharmacokinetic properties. For Karpas-299, a dose of 0.3 mg/kg provided 5/6 cure for cAC10-10 and 5/6 cure for cAC10-1 (Fig. 4). At 0.15 mg/kg, 5/6 cure was observed for cAC10-10 and only 2/6 cure was observed for cAC10-1 (Fig. 4). Thus, for Karpas-299, greater antitumor activity was observed at lower dose levels for cAC10-10, with both ADCs demonstrating high treatment rates above this level.

Schemes 3 and 4 describe the synthesis of analogs of non-PEGylated linker 1 and PEGylated linker 10, respectively, including a Na-maleimido-diaminopropionic (MDpr) acid group as the conjugation point. These two linkers were evaluated in the Karpas299 ALCL and Ramos models of Burkitt lymphoma. For Karpas299, cAC10, conjugates 14 (non-PEGylated) and 16 (parallel PEGylated) were administered once at 0.2 mg/kg and similar inhibition of tumor growth was observed (Fig. 5). In contrast, in the Ramos model, hBU12-16 showed higher antitumor activity compared to hBU12-14 at two different doses. After a single dose of 2 mg/kg, hBU12-16 provided 5/5 cure compared to 0/5 for hBU12-14 (Fig. 6).

Example 10. Synthesis of mDPR-cys(StBu)-PEG ₂ . ₃₆ -OH scaffold for conjugation and mDPR-cys (81Bi)-PEG ₄₈ . ₇₂ -OH framework for conjugation

Scheme 7. Synthesis of MDpr-Cys (81Bi)-PEG ₂ . ₃₆ -OH

- 110 046498

Scheme 8. Synthesis of MDpr-Cys (StBu)—n3r ₄₈ . ₇₂ -OH

Loading 2-chlorotrityl chloride resin:

A polypropylene syringe equipped with a porous polypropylene disk was loaded with 2chlorotrityl chloride resin. A solution of Fmoc-PEG _n -OH (1 eq.) and DIEA (1 eq.) in anhydrous DCM (10 ml/gram resin) was drawn into a syringe. The syringe was sealed with a rubber stopper and shaken for 5 min, then additional DIEA (1.5 eq.) was added. After shaking for another 30 minutes, MeOH (at least 0.8 ml/g resin) was drawn into the syringe to quench the unreacted resin. After shaking for 5 min, the solution was released from the syringe and the resin was washed with DMF (6x5 ml), DCM (6x5 ml) and diethyl ether (6x5 ml). The resin was dried under vacuum conditions.

Loading of Rink's amide resin: To a solution of Fmoc-protected PEG or amino acid (4 eq.) in anhydrous DMF (10 ml/gram resin) was added HATU (4 eq.) and DIEA (8 eq.). The solution was shaken for 5 minutes and taken into a polypropylene syringe equipped with a porous polypropylene disk loaded with Rink amide resin. The reaction mixture was shaken for a minimum of 2 hours and the completion of the reaction was confirmed using the Kaiser test. The resin was washed with DMF (6x5 ml), DCM (6x5 ml) and diethyl ether (6x5 ml) and dried under vacuum.

Deprotection of Fmoc: Fmoc-PEG _n -2-chlorotrityl resin in a polypropylene syringe equipped with a porous polypropylene disk was allowed to swell for 30 minutes with DCM (10 ml/gram resin). The DCM was released from the syringe and the resin was washed with DMF (6x5 ml). The resin was washed with a solution of 20% piperidine in DMF (3x2 min and 1x60 min) with shaking. Completion of the reaction was confirmed using the Kaiser test and the resulting Fmoc-deprotected resin was washed with DMF (6x5 ml), DCM (6x5 ml) and diethyl ether (6x5 ml) and dried under vacuum.

Amino acid binding:

HATU (3 equiv) and DIEA (6 equiv) were added to a solution of Fmoc-protected PEG acid, amino acid or MDpr(Boc)-OH (3 equiv) in anhydrous DMF (10 ml/gram resin). The solution was shaken for 5 min and taken into a polypropylene syringe containing the amino acid 2-chlorotrityl resin with the Fmoc protection removed. The reaction mixture was shaken for a minimum of 2 hours and completion of the reaction was confirmed using the Kaiser test. The resin was washed with DMF (6x5 ml), DCM (6x5 ml) and diethyl ether (6x5 ml) and dried under vacuum.

Removal of IvDde protecting group:

To remove the IvDde protecting group, the peptide resin was washed with a solution of 2% hydrazine in DMF (2x30 min) with shaking. Completion of the reaction was confirmed by the Kaiser test and the resulting unprotected resin was washed with DMF (6x5 ml), DCM (6x5 ml) and diethyl ether (6x5 ml) and dried under vacuum.

Peptide Resin Cleavage:

The final peptides were cleaved from the resin by treatment with TFA in DCM (30% v/v for 2-chlorotrityl resin or 95% v/v for Rink amide resin) for 15 min. After digestion, the solution was left for an additional 60 min to ensure complete removal of the Boc protecting group from the MDpr residue. The resulting solution was evaporated under a stream of nitrogen, and the resulting peptides were analyzed by LC-MS. The peptides were either used crude or preparatively purified.

- 111 046498 phase-phase HPLC followed by LC-MS analysis.

Example 11: Conjugation of a PEGylated scaffold with an antibody and a drug linker.

Scheme 9. Conjugation of a PEGylated scaffold with fully reduced interchain disulfides of an antibody

maleimide hydrolysis

Scheme 10. Deprotection of PEGylated scaffolds for conjugation

Scheme 11. Drug conjugation with PEGylated scaffolds

where X ^R represents the remainder of the precursor releasable complex group X' in the X'-D portion or the remainder of the releasable complex group X in the -XD portion.

Complete reduction of interchain disulfide bonds of the antibody:

To a solution of the antibody at a concentration of approximately 10 mg/ml in PBS containing diethylenetriaminepentaacetic acid (1 mM) and buffered with additional potassium phosphate (100 mM, pH 7.4), 12 equiv was added. tris(2-carboxyethyl)phosphine (TCEP). The solution was subjected to vortex mixing and incubated at 37°C for 1 hour. Complete reduction of interchain disulfide bonds was confirmed by reverse phase chromatography. Additional TCER was added if recovery was incomplete. After reconstitution, the antibody solution was desalted in PBS containing 2 mM EDTA using 3 cycles of dilution and centrifugation at 4000 x g through a 30 kDa MWCO filter. The resulting fully reduced antibody 34 was filtered through

- 112 046498 ril 0.22 µm centrifuge filter and used immediately or stored at -80°C.

Conjugation of maleimide-containing PEGylated scaffold:

To a solution of fully reduced antibody (34) at a concentration of approximately 10 mg/ml in PBS containing EDTA (2 mM) and buffered with additional potassium phosphate (100 mM, pH 7.4), was added 12 molar equivalents of MDpr-PEGp-OH from 5 -20 mM DMSO stock solution. The resulting solution was left at room temperature for 30 minutes. Complete conjugation was confirmed by reverse phase chromatography. Additional PEG reagent was added if conjugation was incomplete. After conjugation, the antibody solution was desalted in PBS using 3 cycles of dilution and centrifugation at 4000xg through a 30 kDa MWCO filter. The resulting PEGylated antibody solution (36 and 37) was filtered through a sterile 0.22 μm centrifuge filter and used immediately or stored at -80°C.

Removal of tert-butylthiol protecting groups from the PEGylated conjugation framework:

To a solution of PEGylated antibody (36 and 37) at a concentration of approximately 10 mg/ml in PBS containing diethylenetriaminepentaacetic acid (1 mM) and buffered with additional potassium phosphate (100 mM, pH 7.4), 20-30 equivalents of TCEP were added. The solution was vortexed and incubated at 37°C for 3 hours. Complete removal of tert-butylthiol protecting groups was confirmed by reverse phase chromatography. Additional TCEP was added and incubation continued at 37°C if recovery was incomplete.

After reconstitution, the antibody solution was desalted in PBS containing 2 mM EDTA using 3 cycles of dilution and centrifugation at 4000xg through a 30 kDa MWCO filter. The resulting solution of unprotected PEGylated antibodies (38 and 39) was filtered through a sterile 0.22 μm centrifuge filter and used immediately or stored at -80°C.

Drug-linker conjugation of maleimide-containing compounds:

12 mol. Eq.v of maleimide-containing drug linker from 5-20 mM DMSO stock solution. The resulting solution was left at room temperature for 30 minutes. Complete conjugation was confirmed by reverse phase chromatography. Additional linker drug was added if conjugation was incomplete. After conjugation, the antibody solution was desalted in PBS using 3 cycles of dilution and centrifugation at 4000xg through a 30 kDa MWCO filter. The resulting PEGylated antibody-drug conjugate solution (40 and 41) was filtered through a sterile 0.22 μm centrifuge filter, analyzed by size exclusion chromatography (SEC) and stored at -80°C.

Example 12: ADCs incorporating PEG in parallel orientation showed low levels of aggregation SEC conjugate assay: Antibody ADC from PEGylated ADC samples (50 μg) was diluted to 1 mg/ml in PBS and 30 μl injection samples were chromatographed on a analytical SEC (TOSOH TSK-gel G3000SW _X L, 7.8 mm ID x 30 cm, 5 µm) using a Waters 2695 HPLC system. Samples were eluted isocratically using 92.5% 25 mM sodium phosphate (pH 6, 8), 350 mM NaCl and 7.5% isopropyl alcohol, at a flow rate of 1 ml/min.

To investigate the effect of PEG length on ADC aggregation, cAC10-MDpr-vcMMAE ADCs loaded with 8 drug moieties per antibody were prepared with or without PEGylated conjugated scaffolds formed using PEG moieties of different sizes. SEC results are shown in Fig. 8. Without the inclusion of the PEGylated conjugated scaffold (cAC10-A), ADC aggregation was 10.4%. The addition of a PEGylated scaffold provides ADCs with lower levels of aggregation. Aggregation decreased with increasing PEG length up to PEG36 (cAC10-D), where the aggregate peak accounted for 2.0% of the total signal. In the case of cAC10-MDpr-vcMMAE, PEG groups longer than PEG ₃₆ (cAC10-D - cAC10-G) did not lead to a further decrease in aggregation.

Structures of drug-linker compounds included in the SEC study: ADCs were conjugated to the antibody via interchain thiols. Antibody-substituted succinimides can exist in their hydrolyzed forms (i.e., a water molecule is added through one rather than both C-N bonds of the succinimide).

- 113 046498

Example 13: ADCs Incorporating PEG in Parallel Orientation Show In Vitro Activity Similar to Their Non-PEGylated Counterparts In Vitro Cytotoxicity of ADCs Made with PEGylated Conjugated Scaffolds MDpr-vcMMAE-based CD30-targeting ADCs were made with and without the addition of a PEGylated conjugation scaffold . Conjugates of compounds A (non-PEGylated), B (PEG ₁₂ ), C (PEG24) and D (PEG ₃₆ ) were tested against the CD30-positive cell lines Karpas 299 and L540cy. Incorporation of PEG and increasing PEG length resulted in little difference in in vitro cytotoxicity (Table 3). Control ADCs (non-PEGylated and PEGylated) prepared with n-ethylaminomaleimide (NAEM) instead of MDpr-vcMMAE (cAC10-H, cAC10-1 and cAC10-J) did not show any activity in this assay, suggesting that the PEGylated scaffolds do not promote in vitro cytotoxicity.

- 114 046498

Table 3. In vitro cytotoxic activity of aHmu-CD30 ADCs prepared with PEGylated conjugated scaffolds; values represent IC ₅₀ values in ng/ml

CD30+ cell lines ADC Quantity drug groups/Ab Karpas 299 L540cy SASYU-A 8 1.7 5, 6 SASYU-V 8 2.2 5 SASYU-S 8 4.2 5, 5 CAC10-D 8 4.3 4 CACIO-NAEM 8 No effect No effect SASYU-N 8 No effect No effect cAClO-I 8 No effect No effect SASYU-J 8 No effect No effect

When compared with the non-PEGylated cAC10-A conjugate having a drug load of 4, the cAC10-A conjugate with a drug loading of 8 had 2-4X in vitro cytotoxicity against Karpas 299 and L540cy; however, Load 8 ADC was not superior to Load 4 ADC in in vivo xenograft models due to the faster clearance of Load 8 ADC (see Example 14).

PEG 24 cAC10 conjugate, cAC10-10, containing an -XD group comprising mc-PABA(gluc)MMAE, which was derived from the linker-drug intermediate of Example 2 and contains the PEG ₂₄ group in a parallel orientation (drug Mb=8) relative to drug, also had higher activity in xenograft models compared to the corresponding 8-loading non-PEGylated ADC (cAC10-1) and the 8-loading ADC containing the PEG ₂₄ group in a sequential orientation (cAC10-4), the latter of which was derived from Linker-Drug intermediate of Example 1 (see FIGS. 1 and 2).

NAEM blocked conjugation scaffolds used as controls: ADCs are conjugated to the antibody via interchain thiols. Antibody-substituted succinimides can exist in their hydrolyzed forms (i.e., a water molecule is added through one rather than both C-N bonds of the succinimide).

- 115 046498

Example 14 ADCs Including PEG in Parallel Orientation Had Better Pharmacokinetics Compared to ADCs Without PEG

Mice were administered a single IV dose of 3 mg/kg of each ADC with a loading of 8 drug/mAb groups. As would be expected, non-PEGylated ADCs prepared with either mc-vcMMAE (K) or MDpr-vcMMAE (A) were cleared from the circulation much faster than the control conjugate prepared with NAEM. The corresponding PEGylated ADCs C and L showed improved pharmacokinetic properties, i.e. slower clearance (Fig. 9).

Additional compounds included in mouse pharmacokinetic studies - ADCs are conjugated to the antibody via interchain thiols. Antibody-substituted succinimides can exist in their hydrolyzed forms (ie, a water molecule is attached through one rather than both CN bonds of the succinimide).

In the second experiment, mice were administered a single IV dose of 3 mg/kg of each ADC with a loading of 8 drug groups/mAb. As in the example described above (Fig. 9), the ADC prepared with non-PEGylated MDpr-vcMMAE A showed accelerated clearance from the circulatory system (Fig. 10). Three ADCs prepared with PEGylated conjugated scaffolds B, C and D showed improved clearance (Figure 10). In this analysis, ADCs prepared with varying PEG lengths, PEG ₁₂ (B), PEG ₂₄ (C), and PEG ₃₆ (D), showed little difference from each other. As might be expected, control conjugates derived from NAEM-capped PEGylated scaffolds (H, I, and J) showed pharmacokinetics closely matching the NAEM-capped antibody (Figure 10).

- 116 046498

Example 15 ADCs Incorporating PEG in Parallel Orientation Showed Improved Pharmacokinetics Compared to ADCs Without PEG

cAC10-based ADCs generated with (B, C, and D) and without (A) PEGylated conjugated scaffolds were analyzed in the L540cy xenograft model. Animals were administered 2 mg/kg (single dose) of each ADC and tumor volume was measured over time. Tumor volume in untreated animals reached 1000 mm ³ on day 25 of the test. The ADC prepared with a non-PEGylated drug-linker compound (A) cured 2 of 5 mice, with an average time of 57.3 days to achieve tumor volumes of 1000 mm ³ in untreated animals (Fig. 11). The ADC prepared with the PEGylated conjugated scaffold combined with PEG 12 (B) showed activity similar to that of the ADC prepared with A (Figure 12). In this case, 1 in 5 animals were cured and it took an average of 68.5 days to reach tumor volumes of 1000 mm ³ in untreated animals. ADC made with the PEG24-containing scaffold (C) showed improvement over A, curing 4 of 5 mice, with the remaining one tumor reaching a volume of 1000 mm ³ on day 53 (Fig. 13).

In a second experiment, hLIV22-based ADCs (the hLIV22 antibody is described in PCT Publication No. WO 2012/078688, which is incorporated herein by reference) targeting the breast carcinoma antigen, LIV-1, were generated with mc-vcMMAE with ( L) and without conjugation scaffolds obtained with PEG ₂₄ (K). Animals were administered 3 mg/kg (single dose) of each ADC. In untreated animals in this trial, the average time to achieve tumor volumes of 1000 mm ³ was 39.2 days. Treatment with hLIV22-K extended this period to 57.6 days, and the pegylated ADC hLIV22-L further extended this average time period to 71.4 days (Fig. 14).

Example 16: ADCs incorporating PEG in parallel orientation and 16 drug moieties per antibody exhibited less aggregation compared to ADCs containing no PEG

To investigate the effect of PEG on the aggregation of ADCs with loading 16, anti-transferrin receptor conjugates were prepared using MDpr-glucuronide-camptothecin as the -X-D group. ADCs were prepared as having standard loadings of 8 (8 drug groups per antibody) or loadings of 16 (16 drug groups per antibody) with or without inclusion of a PEG group. Conjugation was carried out through interchain disulfides. The PEGylated and control non-PEGylated conjugation scaffolds (PEG scaffold A and control scaffold A, respectively) that were used to prepare ADCs with a drug loading of 16 are the following

- 117 046498

PEG frame A

Control frame A.

Antibody-drug conjugates were prepared from n=23 PEG scaffold A, which is an exemplary intermediate of the ligand and control scaffold A as described in Example 11, by (a) contacting the scaffold with an antibody containing thiol groups capable of conjugation addition to the maleimide group of the framework to form antibody-substituted succinimide groups, (b) removing the thiol protecting groups, and (c) contacting the resulting product with -X-D groups, where X is a releasable complex group consisting of a maleimide group and a cleavable group, where the maleimide groups are from X-D are able to react with the free thiol groups obtained from step (b) by conjugate addition under suitable conditions, which converts the maleimide groups present in XD into additional substituted succinimide groups, while avoiding premature hydrolysis of the succinimide groups from the framework of both X-D and (d) hydrolyzing the total substituted succinimides of the linker-drug compound obtained from step (c) by adding a water molecule through one, rather than both, C-N bonds of the succinimide for each of the succinimide groups introduced from MDpr as a maleimide group.

The PEG framework A is covered by the formula

(ie formula VIIIb), where Z' represents an MDpr-containing group, A represents a central lysine residue and the two Lp represent flanking cysteine residues.

Another suitably protected scaffold that provides a conjugate with a drug loading of 16 is the PEG scaffold B having the structure

(PEG framework B), where n has a value of 36. PEG framework B is covered by the formula

- 118 046498

(ie, Formula VIIId) wherein Z' is an MDpr group, t is 1, and AD and L ^P are each cysteine residues.

Without the inclusion of the PEGylated conjugated scaffold, the aggregation level of ADC with a loading of 16 was 22%. Attachment of a PEGylated scaffold that contains a PEG group in a parallel orientation relative to the drug group reduced the level of aggregation to the level at load 8, i.e. 2% aggregate.

Load 8 and load 16 PEGylated anti-transferrin receptor (sOCT9) ADCs containing an -XD moiety consisting of MDpr-PABA(gluc)-Camptothecin were tested against a panel of TfR ⁺ cancer cell lines. In most cases, doubling the drug content increased the effectiveness of the ADC by approximately 2-fold. In some cases, the effectiveness of ADCs increased by 3-10 times or more, even when the drug load was increased by only 2X. Most notably, the load 16 conjugate was active against the colorectal cell line HT-29 (TfR copy number 23K) and the melanoma cell line SKMEL-5 (TfR copy number 21K), whereas the load 8 conjugate was considered inactive (IC50 >1 μM).

Example 17: ADCs loaded with 4 drug groups per antibody with PEG24 in parallel orientation exhibit lower in vivo activity compared to their non-PEGylated counterparts.

For ADCs, when drug loading was reduced to 4 drug groups per antibody, conjugates containing PEGylated glucuronide-MMAE linker 10 were found to have pharmacokinetic properties similar to those of the non-PEGylated conjugate containing linker 1. Accordingly, PEGylation did not provide increased activity in vivo xenograft models.

The Ahtu-CD30 chimeric antibody cAC10 was conjugated to non-PEGylated linker 1 or PEGylated linker 10 at an average loading of 4 drug groups/antibody and was evaluated in L540cy Hodgkin's lymphoma and Karpas 299 anaplastic large cell lymphoma tumor models. For L540cy (Fig. 13), animals were given a single intraperitoneal dose of ADC at 0.5 and 1 mg/kg. At the higher dose of 1 mg/kg, both PEGylated (cAC10-10) and non-PEGylated (cAC10-1) conjugates were equally effective, curing 5 of 6 mice. However, at a lower dose of 0.5 mg/kg, the non-PEGylated linker (cAC10-1) provided longer-lasting tumor growth inhibition with cure in 2 of 6 mice. However, the PEGylated linker (cAC10-10) was less effective, and not a single mouse was cured. Similar results were obtained in the Karpas2 99 xenograft model (Fig. 16).

These data suggest that, while there is no improvement in the pharmacokinetic properties of the conjugate, PEGylation with 24 PEG groups causes very little reduction in in vivo activity. This may result from decreased enzymatic drug release or decreased permeability due to increased conjugate size resulting from PEGylation.

Example 18 ADCs comprising PEGylated glucuronide-containing drug-linker compounds exhibit in vivo activity consistent with the pharmacokinetic properties of the conjugate.

To determine whether there is an optimal PEG size for the combination of glucuronide and MMAE, a series of PEGx linkers covering non-PEGylated, PEG2, PEG4, PEG8, PEG12, PEG24 and branched PEG4-(PEG4) ₃ were prepared and evaluated. Non-PEGylated ADCs cAC10-14 and hBU12-14, presented in table. 4 and 5, respectively, are structurally similar to PEGylated ACDs but do not contain an L ^P group, whereas in the case of PEGylated scaffolds a lysine residue is present.

Initial in vitro studies demonstrated minimal effect of PEG size on activity in most cell lines tested. Ahtu-CD30 and anti-CD19 antibodies, cAC10 and hBU12, respectively, were conjugated at 8 drug/antibody loadings and evaluated against a panel of lymphoma cell lines. The CD30-positive L540cy lymphoma and L428 Hodgkin lymphoma and Karpas 299 anaplastic large cell lymphoma lines were highly sensitive to all cAC10 conjugates regardless of PEG size, as shown in Table 1. 4.

- 119 046498

Table 4. In vitro cytotoxicity - aCD30 conjugates (IC50 in ng/ml)

CD30+ cell lines CD30- ADC ^a PEGx Karpas 299 L540cy L428 R.L. SASYU-14 absence PEG 0.3 3 85 >1000 SASYU-43 PEG2 0.3 2 10 >1000 SASYU-42 PEG4 0.4 3 16 >1000 SASYU-18 PEG8 0.3 2 18 >1000 SASYU-17 PEG12 0.3 2 19 >1000 SASYU-16 PEG24 0.4 3 8 >1000 SASYU-19 PEG4- (PEG4)3 0, 1 1 8 >1000

^a ADCs with 8 drug/Ab loadings

The activity of hBU12 (anti-CD19) conjugates containing PEGx-glucuronide-MMAE linkers, which was determined in a panel of non-Hodgkin's lymphoma cell lines, was more variable, as shown in Table. 5. PEG size had no effect on ADC activity in the Ramos Burkitt lymphoma cell line. However, it appeared that the activity of the conjugate varied as a function of PEG size in the diffuse large B-cell lymphoma cell lines SU-DHL-4, WSU-DLCL-2 and RL. When measured with IC ₅₀ values, there did not appear to be a relationship between PEG size and activity. However, upon closer examination of the dose-response curves, an inverse relationship was found between PEG size and maximum growth inhibition. These data are shown in Fig. 17.

Table 5. In vitro cytotoxicity - aCD19 conjugates (IC50 in ng/ml)

GDI 9+ cell lines CD19- ADC ^a PEGx Ramos SU- DHL-4 WSU-DLCL- 2 R.L. L540cy hBU12- 14 absence PEG 2 22 5 61 >1000 hBU12- 43 PEG2 2 >1000 12 229 >1000 hBU12- 42 PEG4 3 >1000 5 >1000 >1000 hBU12- 18 PEG8 2 16 211 >1000 >1000 hBU12- 17 PEG12 2 >1000 129 >1000 >1000 hBU12- 16 PEG2 4 4 >1000 3 >1000 >1000 hBU12- 19 PEG4- (PEG4)3 2 >1000 247 >1000 >1000

^a ADCs with 8 drug/Ab loadings

The pharmacokinetic properties of the conjugates spanning the PEGx series were assessed as described above. Rats were administered a single intravenous dose of 1 mg/kg of a conjugate consisting of non-binding humanized IgG (h00) carrying MDpr-PEGx-glucuronide-MMAE linkers, with a loading of 8 drug/Ab groups. Plasma samples were obtained at different time points and the total amount of circulating antibody was determined as above. Antibody clearance showed a direct relationship with PEG size, as shown in FIG. 16. PEGylated conjugates with PEG8, PEG12 and PEG24 showed clearance properties from the bloodstream similar to the naked antibody; whereas shorter PEGs and non-PEGylated analogs were cleared more rapidly from the circulatory system.

PEGx linkers were evaluated in vivo in xenograft models. Studies were carried out in models of CD19-positive RL diffuse large B-cell lymphoma and models of CD30-positive L540cy Hodgkin lymphoma. Ahtu-CD19 (hBU12) conjugates spanning non-PEG linkers with PEG4, PEG8, PEG12 and PEG24 were administered once intraperitoneally at 1 and 3 mg/kg once the mean tumor volume reached 100 mm ³ ; the results for the RL model are shown in Fig. 17. At 1 mg/kg, all groups showed only a very slight reduction in tumor growth, and no significant relationship was observed between PEG size and activity. At a higher dose of 3 mg/kg

- 120 046498 conjugates containing no PEG at all and containing PEG4 achieved tumor growth inhibition with tumor enlargement at approximately day 35. In contrast, conjugates with linkers carrying PEG8, PEG12 and PEG24 achieved complete remission at 3 mg/kg, At the same time, in the PEG24 group, tumor growth resumed in 1 out of 5 mice. The increased activity of PEG8, PEG12 and PEG24 at the higher dose compared to PEG4 and non-PEGylated analogues is consistent with the observed pharmacokinetic properties, FIG. 18.

Example 19 Intratumoral delivery of MMAE is consistent with the pharmacokinetic properties of the conjugate.

Mice with CD30-positive L540cy Hodgkin lymphoma, with tumors measuring about 200 ^mm3 , were administered a single dose of 1 mg/kg cAC10 conjugates with a load of 8 drug groups/Ab with mc-glucuronide-MMAE (linker 1), mc-Lys(PEG24 )glucuronide-MMAE (linker 10), maleimido-PEG24-glucuronide-MMAE (linker 4) or MDpr-Lys(PEG24)-glucuronide-MMAE (linker 16). Tumors were taken for study 3 days after administration of the drug, and the intratumoral concentration was determined by mass spectrometry. Consistent with the pharmacokinetic properties of the conjugates, ADCs with PEG24 in a parallel configuration (linkers 10 and 16) delivered significantly greater amounts of MMAE to the tumor compared to the non-PEGylated conjugate (cAC10-1), as shown in FIG. 20. In addition, conjugates containing PEG24 as an expanding group in the series between maleimide and glucuronide (cAC10-4) delivered 4 times less MMAE than the analogue (cAC10-10). Finally, inclusion of mDPR maleimide (cAC10-16) further increased MMAE delivery compared to the maleimidocaproyl-containing analogue (cAC10-10).

Example 20 ADCs loading 8 drug groups per antibody with PEGylated linkers that retain the pharmacokinetic properties of the parent antibody are better tolerated in vivo compared to their shorter PEG and non-PEGylated counterparts.

Balb/c mice (n=3) were intraperitoneally administered a single dose of 50 mg/kg conjugate on day 0. Mice were observed daily for external signs of soreness and body weight was measured; animals were sacrificed if they lost more than 20% of their body weight or if they were considered moribund. The change in body weight from day 0 was plotted as a function of time, FIG. 21. Data plotting was stopped for each group when at least one animal was killed. Mice administered the non-PEG (IgG-14 and -44), PEG2 (IgG-43) and PEG4 (IgG-42) conjugates exhibited significant weight loss or external signs of toxicity and were sacrificed between days 5 and 7. In contrast, mice treated with conjugates containing PEG8 (IgG-18), PEG12 (IgG-17) and PEG24 (IgG-16) showed minimal weight loss and no outward signs of agony. These data, together with the pharmacokinetic profiles shown in FIG. 18 suggest that conjugates with reduced pharmacokinetic exposure showed higher acute toxicity.

Example 21: Maximizing PEG Length

As the length of the PEG chain on the linker drug increases, the overall size and hydrodynamic radius of the conjugate will also increase. This is illustrated in FIG. 22, which shows analytical size exclusion chromatography data showing trace amounts of ADCs prepared with drug-linker compounds 18, 17 and 16 containing 8, 12 and 24 PEG units, respectively. From first principles, as the apparent size of an ADC increases, its ability to diffusively transport in an in vivo system can be expected to decrease. This may have the undesirable effect of reducing the rate or extent to which the ADC can penetrate solid tumors. This reduced diffusive transport capacity can also be observed in plasma pharmacokinetics by fitting the data to a two-compartment model that includes a rate expression for the distribution and elimination phases. Pharmacokinetic data for ADCs prepared with drug-linker compounds 18, 17 and 16 (containing 8, 12 and 24 PEG groups, respectively) were collected over 21 days and fitted to a two-compartment model, and the half-life for these two processes (distribution and elimination) is shown in Fig. 23.

Based on these data, it is obvious that increasing the PEG chain from 8 to 12 units leads to a slower elimination of the substance from plasma (increasing t1/2 by approximately 2 days), but doubling the PEG, from 12 to 24 units, leads to a slight additional improvement of pharmacokinetic properties. Conversely, t1/2 of distribution increases almost linearly over this range, such that doubling the PEG chain from 12 to 24 units nearly doubles the half-life required for distribution into the tissue compartment. These data suggest that 12 PEG units may represent the optimal length for this drug-linker connection, since larger PEGs result in a decrease in the rate of distribution without significantly affecting the rate of elimination. This example shows how pharmacokinetic data can be used to select the optimal PEG size for any given drug linker.

Example 22. Preparation of multiplex PEGylated scaffolds

Schemes 12-14 represent the synthesis of multiplex PEGylated scaffolds A and B, which provide ADCs containing 16 drug moieties/antibody, and multiplex PEGylated scaffold C, the structure of which is shown below, using the peptide coupling methods described for PEGylated scaffolds. providing 4 and 8 drug groups/antibody.

PEG scaffold C covered structure

where Z' is a maleimide-containing group, A is a lysine-lysine branching residue, t is 1, and each AD and each Lp is a cysteine residue.

Scheme 12. Synthesis of MDpr-Cys(StBu)-Ala-Cys(StBu)-ror ₃₆ -QH (framework B)

C.I.

Fmoc-PEG ₃₆ -OH,

1) Piperidine, DMF

2) Fmoc-Cys(StBu)-OH, HATU, DIEA, DMF

1) Piperidine, DMF ₂₇ 2) Fmoc-Ala-OH, HATU, DIEA, DMF

1) Piperidine, DMF

2) Fmoc-Cys(StBu)-OH, HATU, DIEA, DMF

------------------------------n _with ^N

Fmoc

- 122 046498

Scheme 13. Synthesis of Mn3r24-Cys(StBu)-Lys (MDpr)-Cys (8Bi)-PEG24-OH (framework A)

C.I.

Fmoc-PEG ₂₄ -OH, DIEA, DCM

Fmoc^NH

1) Piperidine, DMF

2) Fmoc-Cys(StBu)-OH, HATU, DIEA, DMF

1) Piperidine, DMF

2) Fmoc-Lys(lvDde)-OH, HATU, DIEA, DMF

1) Piperidine, DMF

2) Fmoc-Cys(StBu)-OH, HATU, DIEA, DMF

1) Piperidine, DMF

2) mPEG ₇₄ -OH, HATU, DIEA, DMF

FmoC

Fmoc.

ΗΝ- ^ΙνΜ θ

. IvDde

Fmoc'

HN^ ^IVDde

Scheme 13 (continued). Synthesis of mPEG ₂₄ -Cys(StBu)-Lys(MDpr)-Cys (SBu)-l GEG24-O11

TFA, DCM

1) Hydrazine, DMF

2) MDpr(Boc)-OH, HATU, DIEA, DMF

--------------------------

(Frame A)

- 123 046498

Scheme 14. Branched PEGylated drug-carrier scaffold (scaffold C)

HN OME Fmoc

- 124 046498

MS data for Frames A, B and C, obtained in accordance with schemes 12-14, are presented in table. 6

1) Piperidine, DMF mPEG37-OH, HATU, DIEA, DMF

1) Hydrazine, DMF

2) MDpr(Boc)-OH, HATU, DIEA, DMF

- 125

Table 6. Mass spectrometric data for multiplex PEGylated scaffolds

PEGylated drug scaffold Calculated weight Found weight Branched frame carrier drug (Frame C, where n=37) 4872.5 1624.92 as (M+ZN)/3 MDpr-Cys(StBu)-Ala-Cys(StBu)- PEG36-OH (Framework B, where n=36) 2293.2 1147.85 as (M+2N)/2 mPEG ₂₄ -Sys (StBu) -Lys (MDpr) - Cys (StBu) -PEG ₂₄ -OH (Framework A where n=2 3) 2920.5 1461.32 as (M+2H)/2

Example 23. Preparation of ADCs comprising multiplex PEGylated scaffolds.

Schemes 15-16 show the conjugation of PEGylated scaffolds to an antibody and drug linker. To a solution of fully reduced antibody (34) at a concentration of approximately 10 mg/ml in PBS containing EDTA (2 mM) and buffered with additional potassium phosphate (100 mM, pH 7.4), 12 molar equivalents of a PEGylated branched drug carrier were added from a 5-20 mM DMSO stock solution. The resulting solution was left at room temperature for 30 minutes. Complete conjugation was confirmed by reverse phase chromatography. Additional PEG reagent was added if conjugation was incomplete. After conjugation, the antibody solution was applied to a 1 mL HiTrap MabSelect SuRe column (GE Healthcare Bio-Sciences, Pittsburgh, PA) using a syringe pump and washed with 10 mL of PBS containing EDTA (2 mM) at 1 mL/min. To remove tert-butylthiol protecting groups, the column was washed with 3 ml of 10 mM TCEP buffered with additional potassium phosphate (100 mM, pH 7.4) for 1 h at 37°C. The column was then washed with 10 ml of PBS containing EDTA (2 mM) at 1 ml/min and the purified antibody-scaffold conjugate was eluted with 50 mM glycine (pH 3.0). Protein-containing fractions were pooled and neutralized with 10% (v/v) 800 mM potassium phosphate, 500 mM NaCl, and 500 mM EDTA (pH 7.4). The resulting solution (36) was filtered through a sterile 0.22 μm centrifuge filter and used immediately or stored at -80°C.

To a solution of the unprotected PEGylated antibody (35) at a concentration of approximately 5 mg/ml in PBS containing EDTA (2 mM) and buffered with additional potassium phosphate (100 mM, pH 7.4), 48 molar equivalents of the maleimide-containing drug linker were added from a 5-20 mM DMSO stock solution. The resulting solution was left at room temperature for 30 minutes. Complete conjugation was confirmed by reverse phase chromatography. Additional linker drug was added if conjugation was incomplete. After conjugation, the antibody solution was desalted in PBS using 3 cycles of dilution and centrifugation at 4000x g through a 30 kDa MWCO filter. The resulting PEGylated antibody-drug conjugate solution (37) was filtered through a sterile 0.22 μm centrifuge filter, analyzed by size exclusion chromatography (SEC) and reverse phase chromatography, and stored at -80°C.

- 126 046498

Scheme 15. Conjugation of a maleimide-containing PEGylated branched drug-carrier framework and removal of tert-butylthiol protecting groups

Scheme 16. Conjugation of maleimide-containing drug-linker groups to a branched drug carrier

Example 24. Preparation and biological activity of ADCs containing multiplex PEGylated scaffolds

Auristatin and camptothecin-loaded ADCs were prepared from a PEGylated multiplex scaffold C, where n=37, using the procedures of Example 23. The amount of aggregation was below the level of quantitation, but size exclusion chromatography indicated that MMAE-loaded ADCs could exist in dimeric form.

cAC10 load 32 conjugate containing an -XD group consisting of mc-VC-PABA-MMAE showed >5X improvement in cytotoxicity against L540cy (CD30 copy number 433K) compared to ADC load 8, even though there was only 4X increasing drug load. More significantly, the 32-load conjugate had activity against L-428, which is another Hodgkin's lymphoma cell line, despite the fact that this cell line had a significantly lower copy number of the target antigen (CD30 copy number 77K), then as having a load of 8, the conjugate was considered inactive (IC ₅₀ >1 μM). Also having a 32 MMAE load, the conjugate had cytotoxic activity against the CD30+ multidrug-resistant ALCL cell line. In contrast, the conjugate having a loading of 8 MMAE was considered inactive against

- 127 -

Claims (29)

both multidrug-resistant cell lines, although it had activity that was similar to that of the loading 32 conjugate against the parent cell line. The cAC10 load 32 conjugate, which contains an -X-D group consisting of MDprPAB(gluc)Camptothecin, showed a 3-4X improvement in cytotoxicity against L540cy compared to the load 8 conjugate, but, like the load 8 conjugate, was considered inactive against L-428 . Load 32 conjugate had >5X cytotoxicity against multidrug resistant ALCL cell lines compared to load 8 conjugates. The hBU12 load 32 conjugate, also containing an -X-D group including MDpr-PAB(gluc)Camptothecin, also showed a >5X improvement in cytotoxicity compared to the load 8 conjugate against Raj and Ramos and was active against RL, which has the lowest copy number C19 . In contrast, load 8 conjugate was inactive. Table 7. Mass spectrometry data for ADCs containing multiplex PEGylated scaffolds ADC Calculated weight (light chain, heavy chain) Found mass (light chain, heavy chain) Load=16 MDpr-glucuronide-Camptothecin cOKT9 ADC ⁽¹⁾ 29.092, ND 29.094, ND Load=32 MDpr-glucuronide-Camptothecin sACJ ADC ^(2> 32.501, ND 32.505, ND Load=16 MDpr-VC-MMAE sACYu ADC ⁽¹⁾ 29.104, 66.460 29.108, 66.465 Load=16 mc-VC -MMAE sACYu ADC ⁽³⁾ 28.476, 64.575 28.481, 64.582 Load=32 mc-VC -MMAE sACYu ADC ⁽²⁾ 33.514, 79.690 33.514, 79.691 Load = 32 MDpr-glucuronide- MMAE cACYu ADC ^<2) 33.505, 79.664 33.504, 79.665 Obtained with mPEGgd^fShi)^ (MDpr)-Cys (StBu)-PEF24-OH 2: Prepared with PEG ₃₇ branched backbone drug carrier 3: Prepared with MDpr-Cys(StBu)-Ala-Cys(StBu)-roF36-OH CLAIM 1. A drug-linker compound, wherein the drug-linker compound is represented by the structure of formula IV P.E.G. I/\ Z'---L ^p - A _s Kx - Dl \ A (IV) or a pharmaceutically acceptable salt of such a compound, where D represents a hydrophobic drug group corresponding in structure to a hydrophobic cytotoxic, cytostatic or immunosuppressive drug, having a hydrophobicity measured as SlogP, more than or within 20% of the SlogP value of monomethyl auristatin E or having a SlogP value (a) from 1.5 to 7 or from 2.5 to 7, (b) from 1.5 to 6, from 2 up to 6 or from 2.5 to 6, (c) from 1.5 to 5, from 2 to 5 or from 2.5 to 5, (d) from 1.5 to 4, from 2 to 4 or 2.5 to 4, or (f) from 1.5 to 3, from 2 to 3 or 2.5 to 3; PEG is a polyethylene glycol group, where the polyethylene glycol group has the formula -^-R ²⁰ --(CH ₂ CH ₂ O) _P ---R ²¹ R ²⁰ ---(CH2CH2O)P.--R ²² -( _CH2CH2O ) _P ·---R ²¹ or -%—R ²⁰ --(CH2CH2O)n.-^-R ²² -(CH2CH ₂ O) _n ,-^-R ²¹ where the wavy line indicates the site of covalent addition to L ^P , R ²⁰ represents a PEG addition group, where the PEG addition group is -C(O)-, -O-, -S-, -S(O)-, -NH-, -C(O)O-, -C (O)C1 _- alkyl, -C(O)C1-10alkyl-O-, -C(O)C1-10alkyl-CO2-, - 128 046498 -C(O)C _1-10 alkyl-KH-, -C(O)C _1-10 alkyl^-, -C(O)C _1-10 alkyl-C(O)-KH-, -C(O) )C _1-10 alkyl-KH-C(O)-, -C _1-10 alkyl, -C _1-10 alkyl-O-, -C _1-10 alkyl-CO ₂ -, -C _1-10 alkyl- KH-, -C _1-10 alkyl^-, -C _1-10 alkyl-C(O)-KH-, -C1-10alkyl-KH-C(O)-, -CH2CH2SO2-C1-1oalkyl-, -CH ₂ C(O)-C1-10alkyl-, =N-(O or K)-C1-yualkyl-O-, =N-(O or N) -C _1-10 alkyl-KH-, =N-(O or K)-C _1-10 alkyl-CO ₂ -, =N(O or K)-C _1-10 alkyl^-, R ²¹ is a PEG blocking group, where the PEG blocking group is -C1- _C10 alkyl, -C2 _{- C10} alkyl- _CO2H , _-C2 - _C10 alkyl-OH, _-C2 -C ₁₀ alkyl-MH ₂ , C ₂ -C ₁₀ alkyl-MH (C1-C ₃ alkyl), or C ₂ -C ₁₀ alkyl-K^-C alkyl) ₂ , R ²² is a PEG-linking group for linking multiple chains of PEG subunits together, where the PEG-linking group is -C _1-10 alkyl-C(O)-KH-, -C _1-10 alkyl-NHC(O)- , -C _2-10 alkyl-NH-, -C ₂ -C ₁₀ alkyl-O-, -C1-C ₁₀ alkyl-S- or -C ₂ -C ₁₀ alkyl-NH-, the subscript n independently represents 6 to 72, 8 to 72, 10 to 72, 12 to 72 or 6 to 24; subscript e represents from 2 to 5 and each n' represents from at least 8 or at least 10 to no more than 36, provided that the PEG group contains a total of no more than 72 -CH ₂ CH ₂ O subunits -; Z' is an extension group capable of forming a covalent attachment to an intact antibody or antigen binding fragment of an antibody to provide an antibody-drug conjugate by reaction with a reactive thiol functional group of the antibody or antigen binding fragment of an antibody, where Z' has the formula where the wavy line near R ¹⁷ indicates the site of covalent attachment to LP and R ¹⁷ represents -C1-C ₁₀ alkylene-, C1-C ₁₀ heteroalkylene-, -C1-C ₁₀ alkylene-C(=O)-, C1-C ₁₀ heteroalkylene-C(=O)-, -C ₃ -C ₈ carbocyclo-C(=O)-, -O-(C1-C ₈ alkyl)-C(=O)-, -C1-C ₁₀ alkylene-NH -, C1-C ₁₀ heteroalkylene-NH-, -O(C1-C ₈ alkyl)-NI I-, C1-C ₁₀ alkylene-S-, C1-C ₁₀ heteroalkylene-S-, or -O-(C1- C ₈ alkyl)^-, where R ¹⁷ is optionally substituted with -(CH2)xNH2, -(CH2)xNHR ^a and -(CH2)xNR ^a 2, where the subscript x is an integer having a value of 1-4, and each R ^a is independently selected from the group consisting of C _1-6 alkyl and C _1-6 haloalkyl, or two R ^a groups are combined with the nitrogen atom to which they are bonded to form an azetidinyl, pyrrolidinyl or piperidinyl group; X is a releasable complex group, where the releasable complex group has the formula -|-Q ^c °_qCL—Q ^sp -* where Q ^GO is a covalent addition group, Q ^SP is an optional spacer group, Q ^GL is a cleavable group, asterisk indicates the site of covalent attachment to the drug group, the wavy line indicates covalent attachment to L ^P or A, depending on the presence or absence of A, respectively, the covalent attachment group (Q ^GO ), if present, has the formula each R ¹⁰⁰ is independently selected from the group consisting of -H or -C1-C ₃ alkyl, or selected from the group consisting of -H or CH ₃ , R ¹¹¹ represents p-hydroxybenzyl, methyl, isopropyl, isobutyl, sec-butyl, -CH2OH, - 129 046498 -CH(OH)CH3, -CH2CH2SCH3, -CH2CONH2, -CH2COOH, -CH2CH2CONH2, -CH2CH2COOH, -(CH ₂ ) ₃ NHC(=NH)NH ₂ , -(CH2) ₃ NH2, -(CH2) ₃ NHCOCH ₃ , -(CH2) ₃ NHCHO, -(CH2) ₄ NHC(=NH)NH2, -(CH2)4NH2, -(CH2)4NHCOCH3, -(CH2)4NHCHO, -(CH2)3NHCONH2, -(CH2)4NHCONH2, -CH ₂ CH ₂ CH(OH)CH ₂ NH ₂ , 2-pyridylmethyl-, 3-pyridylmethyl-, 4-pyridylmethyl, each subscript with independently represents an integer from 1 to 10 or from 1 to 3 or Q ^CO has the formula: ;-NH--R ¹³ —C—|R ¹³ is independently selected from -C ₁ -C ₆ alkylene-, -C ₃ -C ₈ carbocyclo-, -arylene-, -C ₁ -C ₁₀ heteroalkylene-, -C ₃ C ₈ heterocyclo-, -C^C^alkylene-arylene-, -arylene-C ₁ -C ₁₀ alkylene-, -C ₁ -C ₁₀ alkylene-(C ₃ -C ₈ carbocyclo)-, -(C ₃ -C ₈ carbocyclo)-C ₁ -C ₁₀ alkylene-, -C ₁ -C ₁₀ alkylene-(C ₃ -C ₈ heterocyclo)- or -(C ₃ -C ₈ heterocyclo)-C _g C ₁₀ alkylene-, the cleavable group is is a peptide cleavable group representing a dipeptide with a cleavable bond to Q ^SP or D, depending on the presence or absence, respectively, of Q ^SP , wherein said cleavable bond is a tumor-associated protease cleavable group, or the peptide cleavable group is a dipeptide -val- cit-, -phe-lys- or -val-ala- having a cleavable bond to ^QSP or D, and the spacer group ( ^QSP ), if present, includes a p-aminobenzyl alcohol (PAB) group that is linked to the dipeptide of the peptide cleaved through the nitrogen atom of the amino PAB group and connected directly to the hydrophobic group of the drug through a carbonate, carbamate or ether functional group, where the PAB group and the adjacent carbonyl group of the carbonate or carbamate group are part of the spacer group; or Q ^SP has the formula -NHb-cl II where Q represents -CpCzalkyl, -O-^-Czalkyl), -halogen, -nitro or -cyano, subscript m is an integer selected from 0-4, and a wavy line near the nitrogen atom indicates the site of covalent attachment to the dipeptide of the peptide cleavable group, and a wavy line near the carbonyl carbon atom indicates the site of covalent attachment to the oxygen or nitrogen atom of the hydrophobic group of the drug; where the split group has the formula Su is a sugar moiety, -O'- is an oxygen glycosidic bond that is cleaved by a glycosidase, or the sugar moiety is a glucuronic acid residue, and the glycosidase is a beta-glucuronidase, each R independently represents hydrogen, halogen, -CN, or - NO ₂ , the wavy line indicates the site of covalent attachment to L ^P or A, depending on the presence or absence of A, respectively, directly or indirectly through the covalent attachment group, and the asterisk indicates the site of covalent attachment of the hydrophobic group of the drug, directly or indirectly through the spacer group where the spacer group, if present, is co6oi-C(=O)-; parallel connecting link (L ^P ) has the formula where the subscript n is an integer from 1 to 4, X ^P is selected from the group consisting of -O-, -NH-, -S- and -C(=O)-, - 130 046498 R ¹ and R ² are independently selected from the group consisting of -H, -C ₁ -C ₃ alkyl, and the wavy line indicates sites of covalent attachment in the drug-linker connection; A is an optional branching group, where the branching group is a single group, is at least trifunctional, or contains 2, 3, 4 or 5 subunits to provide the required number of covalent attachment sites for the LP and X-D groups, where the branching group or subunit thereof is independently selected from amino acid, amino alcohol, amino aldehyde residues, or a combination thereof, wherein at least one of such residues has a side chain functionalized to provide a covalent attachment site to the LP and a covalent attachment site to the X-D group, or wherein the branching group has the formula where R ¹⁰⁰ is as previously defined for LP, each Y' is independently selected from the group consisting of NH, O and S, each subscript c is independently an integer from 1 to 10 or from 1 to 3, where the wavy line indicates two or three of the three covalent attachment sites in a ligand-drug junction; the subscript m is an integer from 1 to 4 or m is 1 or 2 and the subscript s has the value 0 or 1, provided that when the subscript s has the value 0, the subscript m has the value 1, and when the subscript the subscript s has the value 1, the subscript m has the value 2, 3 or 4. 2. The drug-linker compound according to claim 1, where LP is a trifunctional amino acid residue having the structure where the wavy lines indicate sites of covalent attachment in the drug-linker junction. 3. The drug-linker compound according to claim 1, wherein the drug-linker compound has structure IVa or IVb P.E.G. Z'----L ^p ----A---(λ----d) ^λ ' 2 IVa P.E.G. Z'----L ^p ----X----D IVb or a pharmaceutically acceptable salt of such a compound. 4. The drug-linker compound of claim 1, wherein the subscript s is 0, so A is absent. - 131 046498 5. The drug-linker compound according to claim 1, wherein the subscript s has a value of 1 and the subscript m has a value of 2, 3 or 4. 6. The drug-linker compound according to any one of claims 1 to 5, where each LP has structure I 7. The drug-linker compound according to any one of claims 1 to 6, where Z' has the structure optionally protected by an amine protecting group, wherein the wavy line indicates covalent attachment to the remainder of the drug-linker compound structure. 8. The drug-linker compound according to claim 1, wherein the drug-linker compound has the structure or a pharmaceutically acceptable salt of such a compound. 9. The drug-linker compound according to claim 1, wherein the drug-linker compound has the structure - 132 046498 or a pharmaceutically acceptable salt of such a compound. 10. The drug-linker compound according to claim 1, wherein the drug-linker compound has the structure - 133 046498 or a pharmaceutically acceptable salt of such a compound. 11. The drug-linker compound according to claim 1, wherein the drug-linker compound has the structure - 134 046498 or a pharmaceutically acceptable salt of such a compound, wherein R ²¹ is a PEG blocking group and the subscript n is an integer selected from 6 to 72, 8 to 72, or 8 to 24. 12. The drug-linker compound according to claim 1, wherein the drug-linker compound has the structure - 135 046498 or a pharmaceutically acceptable salt of such a compound, wherein R ²¹ is a PEG blocking group and the subscript n is an integer selected from 6 to 72 or 8 to 72 or 8 to 24. 13. The drug-linker compound according to claim 11 or 12, where the subscript n is 8, 12 or 24. 14. The drug-linker compound according to any one of claims 11 to 13, where R ²¹ represents methyl, ethyl or propyl. 15. The drug-linker compound according to claim 1, where the compound is represented by the structure or or a pharmaceutically acceptable salt of such a compound, wherein R ^pr represents hydrogen or a protecting group and n represents 8, 10, 12 or 24. 16. The drug-linker compound according to claim 1, where the compound is represented by the structure - 136 046498 or a pharmaceutically acceptable salt of such a compound. 17. The drug-linker compound according to claim 1, where the compound is represented by the structure or a pharmaceutically acceptable salt of such a compound. 18. The drug-linker compound according to claim 1, where the compound is represented by the structure or a pharmaceutically acceptable salt of such a compound. 19. The drug-linker compound according to claim 1, where the compound is represented by the structure or a pharmaceutically acceptable salt of such a compound. 20. A ligand-drug conjugate compound having a drug-linker component corresponding in structure to the drug-linker compound according to any one of claims 1 to 13, covalently attached to the ligand group by the reaction of a reactive thiol functional group of an antibody or an antigen-binding fragment of an antibody with Z 'drug-linker compounds, wherein the antibody or antigen-binding fragment of an antibody specifically binds to a cancer cell- or tumor cell-associated antigen, wherein said binding represents a drug group of the drug-linker conjugate to a particular population of cancer cells or tumor cells to which the ligand group interacts. 21. The ligand-drug conjugate compound according to claim 20, where the compound is represented by the structure - 137 046498 or a pharmaceutically acceptable salt of such a compound, wherein R ^pr represents hydrogen or a protecting group; p is an integer ranging from 1 to 14 and n is 8, 10, 12 or 24. 22. The ligand-drug conjugate compound according to claim 20, where the compound is represented by the structure or a pharmaceutically acceptable salt of such a compound, and p is an integer ranging from 1 to 14. 23. The ligand-drug conjugate compound according to claim 20, where the compound is represented by the structure or a pharmaceutically acceptable salt of such a compound, wherein p is an integer ranging from 1 to 14. 24. The ligand-drug conjugate compound according to claim 20, where the compound is represented by the structure - 138 046498 or or a pharmaceutically acceptable salt of such a compound, wherein p is an integer ranging from 1 to 14. 25. The ligand-drug conjugate compound according to claim 20, where the compound is represented by the structure or a pharmaceutically acceptable salt of such a compound, wherein p is an integer ranging from 1 to 14. 26. A pharmaceutical composition comprising a population of ligand-drug conjugate compounds, each of which has one or more drug-linker components, corresponding in structure to the drug-linker compound of any one of claims 119, covalently attached to a ligand group, where the ligand group is obtained by a reactive reaction a thiol functional group of an antibody or an antigen binding fragment of an antibody with a Z' of a drug linker compound, wherein the antibody or antigen binding fragment of an antibody specifically binds to a cancer cell- or tumor cell-associated antigen; and a pharmaceutically acceptable carrier, wherein the average number of drug linker components per ligand group in the population is from 1 to 14. 27. The pharmaceutical composition according to claim 26, wherein each group of PEG drug-linker components of the ligand-drug conjugate compounds contains from at least 8 to no more than 24 -CH ₂ CH ₂ O - subunits. 28. The pharmaceutical composition according to claim 26, wherein each group of PEG drug-linker components of the ligand-drug conjugate compounds contains from at least 12 to no more than 24 -CH ₂ CH ₂ O - subunits. 29. The pharmaceutical composition according to any one of claims 26 to 28, wherein the average number of drug-linker components per ligand group in the population is from about 8 to 14. - 139 -