EA040604B1 - PROTEIN-DRUG BINDING CONJUGATES CONTAINING ANTHRACYCLINE DERIVATIVES - Google Patents

Description

The present invention relates to protein-drug-binding conjugates containing anthracycline toxin derivatives.

Introduction

Covalent conjugates of low molecular weight toxins (MW preferably <2500 daltons) with binding proteins, in particular tumor cell-specific antibodies, are a powerful tool to specifically target cancer cells for their destruction. Therefore, such protein-drug conjugates (BPDCs), in particular antibody-drug conjugates (ADCs), are of high medical and commercial interest for the treatment of cancer. In order to develop effective and safe BPDCs or ADCs for cancer treatment, several aspects must be considered: first, the binding protein or antibody must be specific for a given tumor-specific antigen (TSA), which is unlikely or ideally cannot be expressed by normal or healthy individuals. tissue cells. Second, the covalent bond, or connection, between the drug and the binding protein must be functionally stable enough in the bloodstream to prevent undesired release of the toxic payload into the bloodstream, but must be effective in releasing the drug when bound to cancer cells and/or when internalization into cancer cells. Third, the toxic payload must have a toxicity or activity high enough to cause destruction of cancer cells, even if potentially limited amounts of TSA are expressed on cancer cells and therefore only limited amounts of ADC are internalized, or if release of the toxic payload does not occur when sufficiently high efficiency when binding to cancer cells or when internalized into a cancer cell.

While the first aspect of successful cancer targeting with a tumor-specific antigen (TSA) depends on a thorough understanding of the biology of targeting and targeting molecules designed for specific binding, the second and third aspects, regarding optimal linker and toxin payload, generally relate to the effectiveness of conjugates, protein-drug-binding (BPDC) or antibody-drug conjugates (ADC).

All ADCs currently in clinical trials and two FDA approved cancer treatment ADCs, Takeda's anti-CD30 Adcetris® (brentuximab vedotin) ADC and Kadcyla® (trastuzumab emtansine, or T) anti-HER-2 ADC -DM1) from Roche/Genentech (see Perez et al., 2014) are created by chemical conjugation of toxic payloads using maleimide linker chemistry to either primary amino groups of antibody lysine residues or free thiol groups generated by mild reduction intrachain disulfide bridges of an antibody. Chemical conjugation has two limitations: first, maleimide-based chemical linkers have been found to be undesirably unstable in the presence of human serum albumin and thus lead to the release of toxins in the bloodstream of patients treated with maleimide-containing ADCs. linker (see Alley et al., 2008). Secondly, classical chemical conjugation by maleimide linker chemistry results in heterogeneous BPDCs or ADCs because it is not possible to control which amino or thiol groups are conjugated to. Thus, a Gaussian distribution of the number of drugs covalently linked per antibody is obtained such that the conjugated ADCs have a mean drug-antibody ratio (DAR) ranging from 3.5 to 4. However, individual conjugates may not have any drug, attached (DAR=0) to the antibody, or may have up to 8 drugs attached to the antibody (DAR=8) in the case of cysteine conjugates and even more drugs per antibody (>DAR 10) in the case of lysine conjugates. Thus, classical chemically conjugated ADCs are a heterogeneous mixture of different molecules exhibiting different functional properties (see Panowski et al., 2014), which is clearly undesirable from a regulatory point of view when developing ADCs for the treatment of cancer patients.

Therefore, there is a commercial and medical need to provide ADC or BPDC conjugates that are site-specifically conjugated and thus are homogeneous in terms of drug-antibody ratio.

In addition, there is a commercial and medical need to provide ADCs or BPDCs with a more stable drug-to-protein bond that is more stable in the blood stream than traditional conjugates based on maleimide linker chemistry.

There is also a commercial and medical need for an ADC or BPDC that is more effective and has fewer side effects than the ADC or BPDC currently on the market.

The main features of the invention

The present invention solves these problems. The invention provides toxins for use in protein-drug-binding conjugates, as well as an additional novel technology for conjugating these toxins to said binding proteins in a site-specific manner.

- 1 040604 by eliminating the classic maleimide linker chemistry.

The main advantages of these two features will be presented in the description below.

Linker Technology

Both the aforementioned serum instability and the heterogeneity of chemically conjugated and maleimide-linker-containing BPDCs or ADCs are significantly responsible for the safety of these drugs in cancer patients, since both promote non-specific toxin release (drug de-drugging) of such ADCs in patients.

On the one hand, classical maleimide linkers can be degraded by free thiols in human serum, in particular cysteine-34 of human serum albumin, which, as the most abundant serum protein, provides the highest concentration of free thiols in human serum. The cysteine-34 of human serum albumin can cleave the thioether bond of maleimide linkers using the so-called retro-Michael reaction, after which the toxin moves and covalently binds to human serum albumin (HSA). The toxin-HSA conjugate can then distribute the toxin in the circulation or in the body without any tumor selectivity (see Alley et al., 2008).

On the other hand, species with higher DAR in chemically conjugated heterogeneous ADCs are known to have shorter serum half-lives due to the higher hydrophobicity of these ADCs and the tendency to aggregate. Therefore, these higher DAR species undergo faster serum clearance, degradation, and toxin release before these ADCs bind to target positive cancer cells. In addition, higher DAR species are also known to result in faster drug deactivation, as individual conjugation sites have different drug deactivation kinetics depending on the structural context of the amino acid containing the toxin.

The aforementioned liability of chemically conjugated ADCs has hindered the success of developing ADCs to the clinical level, although the concept of delivering a highly active cellular toxin to cancer cells by coupling it to a tumor cell-specific antibody is compelling. Due to toxin-associated side effects, the first ADC to be approved by the FDA in 2000, Mylotarg® (gemtuzumab-ozogamicin) from Pfizer/Wyeth, was withdrawn from the market 10 years after FDA approval (http://www.fda .gov/NewsEvents/Newsroom/PressAnnouncements/ucm216448.htm).

Thus, the responsibility of chemically conjugated ADCs narrows current ADC development efforts to toxins with intermediate cellular toxicity, such as dolastatin/auristatin and maytansine-based tubulin polymerization-inhibiting drugs. In fact, more than 90% of all ADCs currently in clinical evaluation contain monomethylauristatin E (MMAE) or F (MMAF) or maytansine-associated toxins (eg DM1 or DM4) (see Mullard (2013)).

However, drugs that inhibit tubulin polymerization cannot achieve activity below the nanomolar range because tubulin, a component of the cellular cytoskeleton, is a very abundant intracellular target protein, so many drug molecules must diffuse or be transported into the cell to disable intracellular cytoskeleton metabolism. required for cell division and survival. The intermediate activity of tubulin polymerization-inhibiting drugs that can tolerate a degree of drug inactivation of the conjugates, and their specific effects on cell division and mitosis, have made toxins with this specific mode of action the most popular for ADC development.

However, specific tumors with low levels of TSA expression will require much stronger toxins. Thus, newer ADC strategies, still in preclinical development, include toxins with high cellular toxicity and different modes of action, in particular DNA-damaging toxins such as duocarmycins (see Doktor et al. (2014) and pyrrolobenzodiazepines (PBD ) (see Hartley & Hochhauser (2012)).

Anthracycline derivatives

A very interesting class of DNA intercalating toxins for use as payloads for BPDC or ADC are anthracyclines due to their proven clinical validation as chemotherapeutic drugs in the treatment of cancer (see Minotti (2004)). Anthracyclines are red polyketides with potent antitumor activity originally derived from the Streptomyces species. Many derivatives have been described over the past 40 years, including some that are commonly used as a chemotherapy drug for various solid and hematological cancers, such as doxorubicin (also called adriamycin), daunorubicin, epirubicin, idarubicin, or valrubicin. There is even one anti-CD74 ADC in phase I/II clinical trials for multiple myeloma and other hematologic cancers with doxorubicin as a toxic payload, milatuzmab-doxorubicin (see clinicaltrials.gov identifier: NCT01101594).

- 2 040604

All anthracycline-based chemotherapeutics are known to show limited activity on tumor cells as free drugs with an IC ₅₀ in the pmol/ml range on most tumor cells (Crooke and Prestayko, 1981). Despite the example of the first doxorubin-ADC currently being evaluated in clinical trials, the use of traditional anthracyclines as toxic payloads for ADC strategies is likely to remain in question.

About a decade ago, a new anthracycline derivative called PNU-159682 was described as a metabolite of nemorubicin (see Quintieri et al. (2005) Clin. Cancer Res. 11, 1608-1617) cell killing in vitro in the pico-femtomolar range with one ovarian cell line (A2780) and one breast cancer cell line (MCF7) (WO 2012/073217 A1, Caruso et al.). The structure of the anthracycline derivative PNU-159682, as described in the aforementioned prior art documents, is shown in FIG. 2 for reference purposes and with the official anthracycline numbering system for the reactive carbons of the aglycone tetracyclic structure.

Based on the aforementioned limitations of chemically conjugated ADCs, with respect to maleimide linker instability and drug deactivation in higher DAR species in heterogeneous chemically conjugated ADCs, a highly potent anthracycline toxin such as PNU159682 is expected to be very problematic when applied to classical chemical conjugation from - for the release of the toxin into the bloodstream before targeting tumor cells.

Therefore, active toxins, such as PNU-159682, homogeneous ADCs with defined pharmacokinetic properties and increased serum stability, are required to eliminate or minimize side effects from prematurely released toxins in the patient's bloodstream. However, at the same time, specific killing of tumor cells characterized by low expression of the target should still be possible.

Although the use of PNU-159682 as a payload for an ADC generated using classical maleimide linker chemistry approaches has been described previously (WO 2009/099741 A1, Cohen et al.), no functional data has been provided in this prior art document. The first functional data on PNU-159682 bound to antibodies with various linker and spacer structures in the context of chemically conjugated and heterogeneous maleimide-linker-containing conjugates on tumor cells was described in the prior art document WO 2010/009124 A2 (Beria et al.) , but safety and pharmacokinetic data were not provided.

Preferred Embodiments

According to a first preferred embodiment, anthracycline derivative (PNU) conjugates are described which contain PNU-159682 derivatives having no C14 carbon and an attached hydroxyl group of the tetracyclic aglycone structure characteristic of anthracyclines. As a second preferred embodiment, conjugates of anthracycline derivatives (PNU) are described that lack both the C13 and C14 carbons with a carbonyl function at C13 and a hydroxyl group at C14 of the tetracyclic aglycone structure characteristic of anthracyclines.

In these embodiments, the anthracycline derivative (PNU) conjugates comprise an anthracycline derivative PNU-159682 having the following formula (i) or formula (ii):

formula (i) formula (ii)

These conjugates contain on their wavy line a linker structure which may have different elements, XL ₁ -L ₂ -L ₃ -Y, where L1-L3 are linkers and two of L1-L3 are optional, and where X and Y are optional each represent one or more optional linkers.

Both derivatives are markedly different from PNU-159682, which is a metabolite of the anthracycline nemorubicin and was first described by Quintieri et al. (2005).

Both C13 and C14 carbons with their carbonyl function at C13 and hydroxyl group at C14

- 3 040604 are a mandatory structural feature of PNU-159682, which are not part of the derivative conjugates described here.

Surprisingly, and for the first time, PNU derivatives without carbon 14 and the hydroxyl-attached tetracyclic aglycone structure characteristic of anthracyclines have been shown to exhibit cellular toxicity, for example, in site-specifically conjugated antibody-drug conjugates. Preferred embodiments of this are shown in FIG. 3A, 6A and 6B.

In accordance with another embodiment of the invention, there is provided a protein-drug binding conjugate (BPDC) having the following formula:

VR η

VR η

formula (iv) where L _r L ₃ are linkers and two of L^L3 are mandatory, X and Y are each one or more optional linkers, BP is a binding protein, and n is an integer between >1 and <10.

In this construct, multiple linkers can form a single chain that conjugates one toxin to one binding protein, and/or multiple linkers can connect multiple toxins to one binding protein.

Similarly, linkers can conjugate two or more subunits of the same binding protein to two or more toxin molecules.

The optional linker X can be any chemical linker structure known in the art that has been used in ADC to provide specific toxin release upon internalization into cancer cells (see, for example, Ducry & Stump (2010) or McCombs et al. (2015 ))

Some examples of such linkers described in the prior art, which are provided by way of example only and are not intended to be limiting, are shown below.

- 4 040604

H OO hACH ΛΑ' ³⁴ n hydrazone linker disulfide linker ether linker carbamate linker

Val-cit-PAB linker

Linkers L1, L2 and L3 are described below.

The optional linker Y can be any amino acid chain of up to 20 amino acids that provides optimal conjugation of the binding protein to single chain linkers X, L1, L ₂ , L3 or variants thereof, in particular L3.

In addition, linker structures are provided that allow site-specific conjugation of PNU derivatives with suitable binding proteins, for example, and preferably with antibodies. Thus, the derivatives can be used to prepare site-specifically conjugated, homogeneous protein-drug conjugates that can be used in therapeutic applications such as anti-cancer therapy.

According to another preferred embodiment of the anthracycline derivative (PNU), the linker structure contains as L2 an oligoglycine peptide (Glyn) linked to said anthracycline derivative directly or via another L1 linker such that the oligoglycine (Glyn) peptide has a free amino terminus, and where n is an integer between >1 and <21.

In each case, (Gly) _n (also referred to herein as (Gly) _n -NH ₂ or Glyn-segment) is an oligoglycine peptide segment. In one particularly preferred embodiment, n is an integer between >3 and <10, preferably >3 and <6. Most preferably n=5.

As already described herein, the anthracycline (PNU) derivatives described herein are PNU-159682 derivatives either lacking carbon 13 and 14 or lacking only carbon 14 with functional groups attached.

With regard to formula (i), one preferred embodiment is that the oligoglycine peptide (Gly) _n (>1 and <21, preferably n=3 or n=5) is conjugated to an anthracycline derivative using an alkylenediamine linker (NH ₂ - (CH ₂ )m-NH ₂ , m>1 and <11, preferably m=2) which conjugates to the anthracycline derivative via a first carbon 13 amide bond and conjugates to the carboxy-terminus of the oligoglycine peptide via a second amide bond. A preferred compound, PNU-EDA-Gly ₅ , useful for the preparation of site-specific anthracycline derivative (PNU) conjugates, is depicted in FIG. 3A.

With regard to formula (ii), a preferred embodiment is that the oligoglycine peptide (Gly) _n (>1 and <21, preferably n=3 or n=5) is directly connected to the A ring of the PNU derivative (or carbon 9 of the anthracycline aglycone structure ), so that the carbonyl group of carbon 13 is the carboxy-terminus of the glycine peptide linker. A preferred compound, PNU-Gly5, useful for preparing site-specific anthracycline derivative (PNU) conjugates is shown in FIG. 6A.

With regard to formula (ii), another preferred embodiment is that the oligoglycine peptide (Gly) _n (>1 and <21, preferably n=3 or n=5) is conjugated directly to the A ring of the PNU derivative (or carbon 9 of structure anthracycline aglycone), with an alkyleneamine linker - (CH ₂ ) _m -NH ₂ , m>1 and <11, preferably m=2), which conjugates to the carboxy-terminus of the oligoglycine peptide via an amide bond. A preferred compound, PNU-EAGly ₅ , useful for preparing site-specific PNU-derivative conjugates, is shown in FIG. 6b.

Hereinafter, anthracycline derivative conjugates as described above are also referred to as PNU-EDA-Gly _n -NH ₂ , PNU-Gly _n -NH ₂ or PNU-EA-Gly _n -NH ₂ , or, in short form, PNU-EDA-Gly _n , PNU-Gly _n or PNU-EA-Gly _n , respectively, or in the preferred embodiment with 5 glycine residues, PNU-EDA-Gly ₅ , PNU-Gly ₅ or PNU-EA-Gly ₅ , respectively

- 5 040604 of course.

In addition, the invention provides a protein-binding drug conjugate (BPDC) comprising an anthracycline derivative conjugate as described above, and the derivative further comprises a binding protein conjugated to the free amino terminus of an oligoglycine peptide (Glyn) via an additional amide bond .

According to another embodiment of an anthracycline derivative (PNU) or a protein-drug-binding conjugate (BPDC), an oligoglycine peptide (Gly _n ), designated L2, is conjugated to an anthracycline derivative of formula (i) using an alkylenediamine linker, designated L1, and this alkylenediamine linker is conjugated to an anthracycline derivative by a first amide bond, while it is conjugated to the carboxy-terminus of an oligoglycine peptide by a second amide bond, said alkylenediamine linker-oligoglycine peptide conjugate having the following formula (v):

^-NH-(CH2)m-NH-(Gly) _n -NH2 formula (y) where the wavy line shows a compound with an anthracycline derivative of formula (i); m is an integer between >1 and <11, and n is an integer between >1 and <21. Preferably, m is an integer between >2 and <4, most preferably m=2 (ethylenediamine group, EDA).

The alkylenediamine linker is used to allow attachment of the (Gly) _n linker for sortase-conjugation so that coupling can occur via the C-terminus of the (Gly) _n peptide, thereby providing a free N-terminus of the final toxin-linker adduct for sortase-conjugation. It should be understood that any CH2 methylene group in the alkylenediamine linker may be substituted with another stable bond, such as -O-(ether), -S-(thioether), -NH-(amine), or any other alkyl, heteroalkyl, aryl, or heteroaryl group. or any combination thereof to carry out the invention.

According to another embodiment of an anthracycline derivative (PNU) or a protein-drug-binding conjugate (BPDC), the oligoglycine peptide (Gly _n ) is directly linked to the A ring (or carbon 9) of the anthracycline derivative of formula (ii). To illustrate this, see FIG. 6A.

According to another embodiment of an anthracycline derivative (PNU) or a protein-drug-binding conjugate (BPDC), an oligoglycine peptide (Gly _n ) is conjugated to an anthracycline derivative of formula (ii) using an alkylenamine linker designated as L1, and this alkylenamine linker conjugates with the carboxy-terminus of the oligoglycine peptide via an amide bond, said alkyleneamine linker-oligoglycine peptide conjugate has the following formula (vi):

^-(CH ₂ )m-NH-(Gly)n-NH2 formula (vi) where the wavy line shows a compound with an anthracycline derivative of formula (ii) where m is an integer between >1 and <11 and n is an integer between >1 and <11. Preferably m is an integer between >2 and <4, most preferably m=2 (ethyleneamine group, EA).

An alkyleneamine linker is used to allow attachment of the (Gly)n linker for sortase-conjugation so that coupling can occur via the C-terminus of the (Gly) _n peptide, thereby providing a free N-terminus of the final toxin-linker adduct for sortase-conjugation. It should be understood that any CH ₂ methylene group in the alkylenominine linker may be substituted with another stable bond, for example -O-(ether), -S-(thioether), -NH-(amine) or any other alkyl, heteroalkyl, aryl, or heteroaryl group or any combination of them to implement the invention.

In another embodiment of the protein-drug-binding conjugate (BPDC), the L ₃ linker structure contains a peptide motif that results from specific cleavage of the sortase enzyme recognition motif.

As described elsewhere in this invention, as well as in WO 2014140317, the contents of which are incorporated herein by reference, sortases (also called sortase transpeptidases) form a group of prokaryotic enzymes that modify surface proteins by recognizing and cleaving a specific sorting signal containing a particular peptide motif. This peptide motif is also referred to herein as a sortase enzyme recognition motif, a sortase tag, or a sortase recognition tag. Typically, a given sortase enzyme has one or more sortase enzyme recognition motifs that are recognized. Sortase enzymes may be natural or may be genetically engineered (Doerr et al., 2014).

In a preferred embodiment of the protein-drug-binding conjugate (BPDC), said sortase enzyme recognition motif comprises a pentapeptide.

In a preferred embodiment of a protein-drug-binding conjugate (BPDC), said sortase enzyme recognition motif comprises at least one of the following

- 6 040604 amino acid sequences (N-terminus->C-terminus shown):

lpxtg,

LPXSG and/or

LAXTG.

The first two sortase enzyme recognition motifs are recognized by wild-type Staphylococcus aureus sortase A. The second motif is also recognized by the constructed Staphylococcus aureus sortase A 4S9, and the third is recognized by the constructed Staphylococcus aureus sortase A 2A-9 (Doerr et al, 2014). In all three cases, X can be any of the 20 peptidogenic amino acids.

These sortase enzyme recognition motifs are, for example, fused to the C-terminus of the binding protein, or a domain or subunit thereof, by genetic fusion and co-expressed with it. Said fusion may be performed directly or indirectly via an additional Y linker described elsewhere herein.

It should be noted that after integration into the linker structure and conjugation with L ₂ , L ₃ does not contain the 5th amino acid residue (C-terminus G) of sortase enzyme recognition motifs. In table. 1, the indicated C-terminus of G is shown in brackets. In the case where the sortase enzyme recognition motif is a pentapeptide, L ₃ is thus a tetrapeptide.

Prior to sortase conjugation, sortase enzyme recognition motifs may additionally contain other tags such as His tags, Myc tags or Strep tags (see FIG. 4A in WO 2014140317, the contents of which are incorporated herein by reference), fused to C -end with sortase enzyme recognition motifs. However, since the peptide bond between the 4th and 5th amino acids of the sortase enzyme recognition motif is cleaved after sortase-mediated conjugation, these additional marks will eventually be removed from the fully conjugated BPDC.

Sortase enzyme recognition motifs can be conjugated to a (Gly) _n linker, which conjugates to an anthracycline derivative using the sortase technology described here and in WO 2014140317. During the conjugation process, one glycine residue is released from the (Gly) _n linker.

It should be noted that although these three peptide stretches are shown above in the classical N-terminus->C-terminal direction, the L residue is a residue fused to the C-terminus of the binding protein or the C-terminus of the Y linker by a peptide bond. The 5th amino acid residue (G) from L ₃ is removed by conjugation to the peptide (Gly) _n while the 4th T or S amino acid residue from L ₃ is the residue that actually conjugates to the N-terminus of the peptide (Gly) _n .

Thus, the following table provides an overview of the preferred embodiments of the drug-binding protein conjugate (BPDC) according to the invention, indicating L1-L3.

Table 1. Typical linker structures.

Toxin Li L2 L ₃ (shown here C'->N') Binding protein formula (i) Alkylenediamine group (Gly)n (G)TXPL (G)SXPL (G)TXAL antibody formula(s) Alkylenamine group (Gly)n (G)TXPL (G)SXPL (G)TXAL antibody

As mentioned, it should be noted that after integration into the structure of the linker and conjugation with L2, L ₃ does not contain the 5th amino acid residue (C-terminus of G). Thus, in table. 1 the specified Terminal G is shown in brackets.

According to another embodiment of a drug-binding protein conjugate (BPDC), an anthracycline derivative (PNU) is conjugated by one or more linkers to the carboxy-terminus of the binding protein or to the carboxy-terminus of its domain or subunit.

In another preferred embodiment, n in the oligoglycine (Gly _n ) peptide linker is an integer between >3 and <11, more preferably between >3 and <7, preferably n=3 or n=5. Most preferably, the n in the oligoglycine (Gly _n ) peptide linker is 5.

In one preferred embodiment, the payload is the payload of formula (i).

In a second preferred embodiment, the payload is the payload of formula (ii).

- 7 040604

According to another embodiment of a protein-drug-binding conjugate (BPDC), the binding protein is conjugated to the free amino terminus of an oligoglycine peptide (Glyn) via an amide bond.

According to another embodiment of a drug-protein-binding conjugate (BPDC), the binding protein is at least one selected from the group consisting of:

an antibody, a modified antibody format, an antibody derivative or fragment that retains the properties of binding to a target, an antibody-based binding protein, an oligopeptide binder, and/or an antibody mimetic.

The term binding protein used in this context is equivalent to the term immunoligand used in other publications by the authors, including Appendix 1, which contains additional technical data, description, and implementation regarding sortase enzyme conjugation technology.

Antibodies, also synonymously called immunoglobulins (Ig), usually contain four polypeptide chains, two heavy (H) chains and two light chains (L) and are therefore multimeric proteins or their equivalent Ig homologue (e.g., camelid nanobodies, which contain only single chain heavy chain antibodies (dAbs) that can be derived from either the heavy or light chain); including full-length functional mutants, variants or derivatives thereof, including, but not limited to, murine, chimeric, humanized and fully human antibodies that retain the essential characteristics of epitope binding of the Ig molecule and include immunoglobulins, dual specific, bispecific, multispecific immunoglobulins, as well as immunoglobulins with double variable domain; immunoglobulin molecules can be of any class (eg, IgG, IgE, IgM, IgD, IgA, and IgY) or subclass (eg, IgG1, IgG2, IgG3, IgG4, IgA1, and IgA2) and allotype.

Provided that the binding protein is an antibody, the protein-drug binding conjugate is an antibody-drug conjugate (ADC).

Hereinafter, the ADC conjugates according to the present invention are also referred to as PNU-EDAGly _n -Ab, PNU-Gly _n -Ab or PNU-EA-Gly _n -Ab.

The term antibody-based binding protein as used herein can be any protein that contains at least one antibody-derived VH, _VL , or CH immunoglobulin domain in the context of other non-immunoglobulin or non-antibody-derived components. Such antibody-based proteins include, but are not limited to, (i) Fc fusion proteins of binding proteins, including receptors or receptor components with all or part of the immunoglobulin CH domains, (ii) binding proteins in which the VH and or VL domains are fused to alternative molecular scaffolds, or (iii) molecules in which immunoglobulin V _H and/or V _L and/or C _H domains are combined and/or assembled in a manner not normally found in natural antibodies or antibody fragments.

The term antibody derivative or fragment as used herein refers to a molecule containing at least one polypeptide chain derived from an antibody, not full length, including but not limited to (i) a Fab fragment, which is a monovalent fragment consisting of variable light (VL), variable heavy (VH), constant light (C _L ) and constant heavy 1 (CH1) domains; (ii) an F(ab') ₂ fragment, which is a bivalent fragment containing two Fab fragments linked by a disulfide bridge in the hinge region; (iii) the heavy chain region of the F _ab (F _d ) fragment, which consists of the VH and CH1 domains; (iv) a variable fragment of a fragment (F _v ) that consists of the V _L and VH domains of a single arm of an antibody, (v) a domain fragment of an antibody (dAb) that contains a single variable domain; (vi) an isolated complementarity determining region (CDR); (vii) a single chain F _v fragment (scF _v ); (viii) a diabody, which is a bivalent bispecific antibody in which the V _H and V _L domains are expressed on the same polypeptide chain, but using a linker that is too short to allow pairing between the two domains on the same chain, thereby forcing domains to pair with complementarity domains of another strand and create two antigen-binding sites; and (ix) a linear antibody that contains a pair of tandem F _v segments (VH-CH1VH-CH1) which, together with complementary light chain polypeptides, form a pair of antigen-binding regions; and (x) other non-full length immunoglobulin heavy and/or light chain regions or mutants, variants or derivatives thereof, alone or in any combination. In either case, said derivative or fragment retains its target-binding properties.

The term modified antibody format as used herein encompasses scFv polyalkylene oxide modified antibody drug conjugates, monobodies, diabodies, camelid antibodies, domain antibodies, bi- or trispecific antibodies, IgA, or two

- 8 040604 IgG structures linked by J-chain and secretory component, shark antibodies, American primate framework + non-American primate CDRs, IgG4 antibodies with hinge removed, IgG with two additional binding sites constructed in CH3 domains, re-regioned antibodies Fc to increase affinity with Fc-gamma receptors, dimerized constructs containing CH3+VL+VH, and the like.

The term antibody mimetic as used herein refers to proteins not belonging to the immunoglobulin family, and even to non-proteins such as aptamers or synthetic polymers. Some types have an antibody-like beta sheet structure. Potential advantages of antibody mimetics or alternative scaffolds over antibodies are better solubility, higher tissue penetration, higher heat and enzyme resistance, and relatively low manufacturing costs.

Some antibody mimetics can be provided in large libraries that offer specific binding candidates for all possible targets. As with antibodies, targeted specific antibody mimetics can be designed using high throughput screening (HTS) technologies as well as established display techniques such as phage display, bacterial display, yeast or mammalian display. Antibody mimetics currently developed include, for example, ankyrin repeat proteins (called DARPins), C-type lectins, S. aureus A-domain proteins, transferrins, lipocalins, fibronectin type III domains 10, Kunizza domain protease inhibitors, ubiquitin-derived binders (called affilins), gamma-crystallin-derived binders, cysteine knots or nottins, thioredoxin A backbone binders, SH-3 domains, stradotel, membrane receptor A domains, stabilized by disulfide bonds and Ca2+, CTLA4-based compounds, Fyn SH3, and aptamers (peptide molecules that bind to specific target molecules).

The term oligopeptide binder as used herein refers to oligopeptides that have the ability to bind with high affinity to a given target. The term oligo refers to peptides that contain 5 to 50 amino acid residues.

According to another embodiment of a drug-protein-binding conjugate (BPDC), the binding protein binds to at least one entity selected from the group consisting of a receptor, an antigen, a growth factor, a cytokine, and/or a hormone.

This list defines the different types of targets that a binding protein can bind to. The term receptor as used herein means a cell surface molecule, preferably a cell surface molecule, which (i) binds to specific signaling molecules or groups of specific signaling molecules (i.e. a receptor such as, for example, the VEGF receptor) and/or (ii) has no known ligand (ie, an orphan receptor such as, for example, HER2/neu). Natural receptors are expressed on the surface of a population of cells, or they simply represent the extracellular domain of such a molecule (regardless of whether such a form exists in nature or not) or a soluble molecule that performs a natural binding function in plasma, or within a cell or organ. Preferably, such a receptor is a member of a signaling cascade that is involved in a particular pathogenic process (e.g., a receptor that belongs to the growth factor signaling cascade), or is expressed on the surface of a cell or particle that is involved in a pathological process, such as a cancer cell.

The term antigen as used herein means a substance that has the ability to induce a specific immune response and may include surface proteins or protein complexes (eg, ion channels). Often, antigens are associated with pathogenic entities, such as a cancer cell.

The term cytokine, as used herein, refers to small cell signaling protein molecules that are secreted by numerous cells and are a category of signaling molecules widely used in cell-to-cell communication. Cytokines can be classified as proteins, peptides or glycoproteins, the term cytokine encompassing a large and diverse family of regulators produced throughout the body by cells of diverse embryological origins.

The term growth factor, as used in this context, refers to natural substances capable of stimulating cell growth, proliferation and cell differentiation. Typically, the growth factor is a protein or a steroid hormone. Growth factors are important in regulating various cellular processes.

The term hormone, as used in this context, refers to the chemical secreted by

- 9 040604 my cell, gland or organ in one part of the body that sends messages that affect cells in other parts of the body. The term includes peptide hormones, lipid- and phospholipid-derived hormones, including steroid hormones and monoamines.

In the event that the binding protein binds to a receptor or antigen, a protein-drug-binding conjugate (BPDC) can, for example, be directed to a specific site, such as a pathogenic entity, such as a cancer cell, where a payload, such as a toxin, is delivered. Thus, the systemic toxicity of the toxin or chemotherapeutic agent is reduced, while the local concentration of the latter at the site of action is increased, thereby increasing efficacy while reducing side effects. In addition, the corresponding signaling cascade can be inhibited by the binding of the binding protein. If the payload is a marker, then the latter can thus be used to mark a specific site, such as a cancer cell, characterized by a given surface antigen detected by the binding protein, for diagnosis.

In the event that the binding protein binds to a growth factor, cytokine and/or hormone, a protein-drug-binding conjugate (BPDC) can, for example, be directed to the site to which the cytokine or growth factor hormone normally binds in order to deliver the payload site- in a specific way. In addition, the corresponding signaling cascade can be inhibited by the binding of the binding protein.

The term "bind" as used herein means a well-understood interaction or other non-random association between a binding protein, such as antibodies or antibody fragments, and their targets. Preferably, such a binding reaction is characterized by high specificity and/or sensitivity to the target. Preferably, the coupling reaction is characterized by a dissociation constant (Kd) <10'3 M, preferably <^10'4 M, <10'5 M, <^10'6 M, <^10'7 M, <^10'8 M, <10' ⁹ M and most preferably <10' ¹⁰ .

According to another embodiment, the binding protein has at least two subunits.

In this embodiment, one subunit can be conjugated to the anthracycline derivative PNU-159682 described herein (see FIGS. 3A and 6A and 6B).

Preferably, at least two different drugs can be conjugated to at least two subunits in a site-specific manner. This option provides a versatile toolkit with which a large number of different drug-binding protein constructs can be created.

Preferably, the at least two different drugs are drugs that interfere with different cellular pathways. This means that following the anthracycline derivative conjugate described here, a second toxin can be conjugated to a different subunit of the same binding protein.

Such an embodiment can be implemented, for example, by conjugating two different drugs to each of the two full-length antibody light chains and two full-length antibody heavy chains, respectively, using two different sortase enzymes recognizing different recognition motifs (sortase tags), plus an antibody that contains various C-terminal modifications on the heavy and light chains containing the respective recognition motifs for said different sortase enzymes.

Thus, an antibody-drug conjugate can be created that consists of each two full-length Ig light chains and Ig heavy chains containing different payloads covalently attached to said heavy and light chains.

Such an embodiment preferably results in site-specific conjugation of at least two subunits to form protein-drug-binding conjugates by site-specific, equal payload conjugation to each of said subunits.

In one embodiment of a drug-protein-binding conjugate (BPDC), the binding protein binds to HER-2. Preferably, the binding protein is an antibody specific for HER-2.

In this embodiment, the HER-2 specific antibody is preferably:

a) contains CDRs 1-6 of trastuzumab (humanized hu4D5),

b) contains the heavy chain variable domain and the light chain variable domain of trastuzumab,

c) has an amino acid sequence identity of 90% or greater with regions or domains from a) or b),

d) is trastuzumab or a target-binding fragment or derivative thereof and/or

f) competes with trastuzumab for binding to Her-2.

The anti-HER-2 monoclonal antibody trastuzumab binds to domain IV of the HER-2 antibody.

- 10 040604

Preferably, the aHTu-HER-2 antibody contains the primary amino acid sequences of the IgH and IgL chains, FIG. 11A (Seq ID Nos 1 and 2).

Trastuzumab sequences are also provided in the Drug Database Access Number DB00072 (BIOD00098, BTD00098), which is incorporated herein by reference, and in the IMGT database (VH: http://www.imgt.org/3Dstructure-

DB/cgi/details.cgi?pdbcode=7637&Part=Chain&Chain=7637H & VL: http://www.imgt.org/3DstructureDB/cgi/details.cgi?pdbcode=7637&Part=Chain&Chain=7637L).

In another embodiment of the drug-binding protein conjugate (BPDC), the binding protein binds to CD30. Preferably, the binding protein is an antibody specific for CD30.

In this embodiment, the antibody is preferably:

a) contains CDR regions 1-6 of brentuximab (chimeric cAc10),

b) contains the heavy chain variable domain and the light chain variable domain of brentuximab,

c) has an amino acid sequence identity of 90% or greater with regions or domains from a) or b),

d) is brentuximab or a target-binding fragment or derivative thereof and/or

e) competes with brentuximab for binding to CD30.

The sequences of brentuximab (clone cAc10), which is an antibody component of the approved drug adcetris/brentuximab vedotin [Adcetris/Brentuximab vedotin], are presented in US2008213289A1.

Preferably, the anti-CD30 antibody contains the primary amino acid sequences of the IgH and IgL chains, FIG. 11 (Seq ID Nos 3 and 4).

Preferably, in these embodiments, the toxin is a toxin of formula (i),

L ₁ is an ethylenediamine linker,

L ₂ is an oligoglycine (Glyn) peptide linker (with n having a preferred length of 5 amino acids) and

L ₃ is amino acid residues 1-4 of the sortase-tagged pentapeptide motif (i.e., lacking the C-terminal residue G (5th amino acid residue) that is removed by sortase-mediated conjugation to the peptide (Gly) _n , linker X is absent and Y is a 5 amino acid linker between the C-terminus of the Ig light chain and L ₃ , preferably with the amino acid sequence GGGGS.

Alternatively, the toxin is a toxin of formula (ii), while

L1 is an ethyleneamine linker,

L2 is an oligoglycine (Glyn) peptide linker (with n having a preferred length of 5 amino acids),

L3 is amino acid residues 1-4 of the sortase-tagged pentapeptide motif (i.e., lacking the C-terminal residue G (5th amino acid residue) that is removed by sortase-mediated conjugation to the (Gly)n peptide, linker X is absent and

Y is a 5 amino acid linker between the C-terminus of the Ig light chain and L ₃ , preferably with the amino acid sequence GGGGS.

Furthermore, the present invention provides a method for producing a protein-drug-binding conjugate (BPDC) as described above, wherein the binding protein containing the sortase enzyme recognition motif is conjugated by the sortase enzyme to at least one anthracycline derivative conjugate that contains as L2 an oligoglycine peptide (Glyn).

Sortase technology, its advantages (site-specific conjugation, stoichiometrically determined relationship between toxin and binding protein, high conjugation efficiency) are explained in detail in WO 2014140317 A1, the content of which is incorporated herein by reference. Additional explanations regarding sortase labels are provided above.

A preferred embodiment of the present invention is the conjugation of PNU-derived payloads using SMAC technology to the C-terminus of binding proteins and preferably the C-terminus of antibody or immunoglobulin chains to at least one Ig light chain or Ig heavy chain. This is achieved by designing mammalian cell expression structures to bind protein or immunoglobulin subunits that encode the C-terminal pentapeptide recognition motif to sortase enzymes immediately after the C-terminus of the binding protein or polypeptide subunit of a multimeric binding protein, such as an antibody.

It should be understood that the pentapeptide motif of sortase A from Staphylococcus aureus, which is LPXTG or LPXSG and which was mentioned earlier, is given only as a non-limiting example and can be replaced by any other pentapeptide motif recognizable

- 11 040604 sortase enzymes from other species or other classes, such as sortase B from Staphylococcus aureus, which recognizes the NPQTN pentapeptide motif. Recognition motifs that are recognized by engineered sortase enzymes can also be used, such as LAETG recognized by the engineered version of sortase A from Staphylococcus aureus recently described by Dorr et al. (2014).

WO 2014140317 also provides technical details, description and implementation regarding sortase-conjugation technology, which is also called SMAC technology (sorta-mediated antibody conjugation technology). This technology allows two entities to be conjugated, one labeled with a (Gl y _n )-segment (as mentioned above for a toxin) and one with a so-called sortase label, which is a peptide label that can be attached to, for example, a binding protein.

These sortase tags are oligopeptides, usually pentapeptide motifs, which are fused to a first entity (here: binding protein) to be conjugated to a second entity (here: an anthracycline derivative) such that the C-terminus of said sortase tag oligopeptides remains free . As described in WO 2014140317, this can be achieved by expressing binding proteins from expression vectors encoding additional amino acids per pentapeptide sortase tag.

Such a sortase tag is, for example, LPXTG or LPXSG (for sortase A from Staphylococcus aureus), LPXSG (for engineered sortase A 4S9 from Staphylococcus aureus described in Dorr et al., 2014) or LAXTG (for engineered sortase A 2A9 from Staphylococcus aureus described in Dorr et al., 2014), where X is any of the 20 natural amino acids. However, such sortase marks may differ sequentially in sortase enzymes from other bacterial species or in sortase classes, as described in WO 2014140317 and in the prior art (Spirig et al., 2011).

The second entity contains a glycine segment (Glyn segment) with a free N-terminus (-NH ₂ ), the Glyn segment being an oligoglycine peptide. Preferably n is an integer between >1 and <21. In one particularly preferred embodiment, n is an integer between >3 and <10, preferably n=3 or n=5. Most preferably n=5.

The sortase enzyme is further capable of fusing two entities to each other via a transpeptidation reaction in which the C-terminal amino acid residue (eg, G in LPXTG) is cleaved off and then replaced with the first glycine from said glycine moiety.

In another preferred embodiment, the pentapeptide recognition motif can be directly attached to the last naturally occurring C-terminal amino acid of immunoglobulin light chains or heavy chains, which in the case of human kappa light chain immunoglobulin is the C-terminal cysteine residue, which in the case of human immunoglobulin light chain lambda is a C-terminal serine residue, and which in the case of a human IgG1 immunoglobulin heavy chain may be the C-terminal lysine residue encoded by the human Fcy1 cDNA. However, another preferred embodiment is also to directly attach the sortase pentapeptide motif to the second last C-terminal glycine residue encoded by the human Fcy1 cDNA, since typically the lysine-terminal residues of antibody heavy chains are cut off by post-translational modification in mammalian cells. Consequently, in more than 90% of cases, native human IgG1 lacks the C-terminal lysine residues of IgG1 heavy chains.

In another preferred embodiment, the pentapeptide recognition motif can be attached to the C-terminus of a human IgG1 immunoglobulin heavy chain, where the C-terminal lysine residue encoded by the human Fcy1 cDNA is replaced with an amino acid residue other than lysine.

We have previously described that in some cases (e.g. at the C-terminus of the kappa of Ig light chains, (Beerli et al., 2015) it is beneficial to add additional amino acids between the C-terminus of the binding protein and the sortase tag, L ₃ . loads of conjugation of sortase enzymes to the binding protein In the case of Ig kappa light chains, it was found that adding 5 amino acids (GGGGS) between the last C-terminal cysteine amino acid of the Ig kappa light chain and the sortase tag improved conjugation kinetics so that the C-terminus of the kappa lung Ig chains and Ig heavy chains could be conjugated with similar kinetics (see Beerli et al. (2015). Therefore, another preferred embodiment optionally includes a Y amino acid linker in the range >1 and <21 between the last C-terminal amino acid of the binding protein or subunit antibodies and sortase label L3.

In addition, the present invention provides the use of a protein-drug-binding conjugate (BPDC) according to the above description or obtained using the above method for the treatment of a human or animal suffering from a given pathological condition, at risk of developing a given pathological condition and / or in said person or an animal diagnosed with a given pathological condition.

- 12 040604

In addition, the present invention provides the use of a protein-drug-binding conjugate as described above for the manufacture of a medicament for the treatment of a human or animal suffering from a given pathological condition, at risk of developing a given pathological condition and/or said human or animal is diagnosed with this pathological condition. state.

Preferably, the pathological condition is a neoplastic disease. More preferably, the neoplastic disease is a cancer that has a HER-2 expression score of 1+, 2+, or 3+ as determined by IHC or ISH, and the cancer is preferably a breast cancer, a cancer that is CD30 positive, as determined by IHC, ELISA or flow cytometry, preferably lymphoma, more preferably Hodgkin's lymphoma (HL) or systemic anaplastic large cell lymphoma (sALCL).

HER-2 status can be determined, for example, according to ASCO/CAP guidelines, which are described in Wolff et al 2013.

CD30 status can be determined, for example, according to the method outlined in Young 2014.

The invention further provides a pharmaceutical composition comprising a protein-drug-binding conjugate (BPDC) as described above, or prepared as described above, and at least one other pharmaceutically acceptable ingredient.

Detailed description

To overcome the main limitations of traditional maleimide linker chemistry in relation to the creation of BPDC and ADC conjugates, the authors have previously developed an enzymatic approach to create BPDC or ADC using sequence-specific transpeptidase enzymes either using sortase enzymes or so-called cleaved inteins (see WO 2014140317 A1). In particular, it has been possible to demonstrate that site-specific conjugation of small molecular weight payloads by sortase enzymes, in the context of antibodies, referred to as SMAC technology (Sortase-Mediated Antibody Conjugation Technology), results in ADCs that are equally active as chemically conjugated ADC in killing cancer cells in vitro if the same binding protein and the same payload are used. In addition, SMAC-generated ADCs specific for the HER-2 target resulted in similar strong tumor regression in xenotransplantation models when the same target antibody (anti-HER-2 trastuzumab) and the same toxic payload were used ( DM1) (WO 2014140317 A1). However, firstly, the ADC conjugates created by SMAC did not use maleimide linker chemistry, and secondly, the conjugation reaction was carried out in a site-specific manner with the C-terminus of either IgH or IgL antibody chains, so that more homogeneous ADCs.

In the case of using SMAC technology, site-specific conjugation can be carried out, for example, using a recombinant sortase A enzyme from Staphylococcus aureus, which specifically recognizes the LPXTG or LPXSG pentapeptide motif (X=any of the 20 natural amino acids) and which can be attached to the recombinant antibody for conjugation. Sortase A then uses the oligoglycine stretch as a nucleophile to catalyze the transpeptidation reaction by which the amino group of the oligoglycine makes a nucleophilic attack on the peptide bond between the threonine or serine and the glycine of the LPXTG or LPXSG pentapeptide motif. This results in the cleavage of this peptide bond and the formation of a new peptide bond between the N-terminal glycine of the oligoglycine peptide (see FIG. 1), i.e. causes transpeptidation.

Although trastuzumab-DM1 conjugates generated by cultivar-mediated conjugation have been shown to have comparable potency to chemically conjugated DM1 conjugates (T-DM1 or Kadcyla®, which are already in clinical use), the higher activity of ADCs generated by SMAC technology not achieved (WO 2014140317 A1). This was not expected, since the same target antibodies and the same payload were used.

Based on this and other experiments with various monoclonal antibodies that specifically bind to other TSAs that are potentially expressed at lower levels on cancer cells than the HER-2 target, or that are potentially less efficiently internalized when bound to ADCs (data not shown), it appears that toxic payloads with higher activity than maytansins and/or with a potentially different mode of action are required to obtain sufficiently effective ADCs. In addition, the payload must be modifiable in at least one reactive group, allowing the addition of an oligoglycine peptide to allow sortase-conjugation of the payload with LPXTG- or LPXSG-modified binding proteins. Finally, when using toxins with higher activity, the modification should lead to a stable connection between the glycine segment and

- 13 040604 payload to prevent undesired release of the toxic payload into the bloodstream, but at the same time, the toxin must still cause effective killing of cancer cells by specifically binding and internalizing the BPDC or ADC into the tumor cells.

According to an empirical evaluation of various toxic payloads described in the prior art in the context of SMAC technology, it was concluded that a highly active anthracycline derivative of nemorubicin, called PNU-159682 (Quintieri et al., 2005) (see also FIG. 2) modified with an ethylenediamine spacer to allow for the addition of a pentaglycine moiety could be conjugated very efficiently with modified LPXTG antibodies by SMAC technology yielding almost fully conjugated ADCs based on product analysis by HIC (Hydrophobic Interaction Chromatography) and reversed. -phase chromatography (data not shown). In addition, when this modified derivative of PNU159682, referred to as PNU-EDA-Gly ₅ , is SMAC-conjugated with various monoclonal antibodies, as described in the examples below, TSA-dependent killing of tumor cells with high activity is performed. In particular, it was possible to effectively destroy in vitro human breast cancer cells with low expression of HER-2 using PNU-EDA-Gly ₅ conjugates conjugated by SMAC technology, while maytansine-toxin conjugates gave a very weak effect. This demonstrates the potential utility of a PNU-EDA-Gly ₅ derivative to create active BPDCs and ADCs, preferably containing PNU-EDA-Gly ₅ , or any PNU derivative with an oligoglycine peptide having at least two glycines attached to it. In addition, this demonstrates the benefit of BPDCs and ADCs containing preferably PNU-EDA-Gly ₅ or any PNU derivative with an oligoglycine peptide as a payload for the treatment of oncological diseases.

Although the anthracycline derivative PNU-159682 (FIG. 2) and its use in the context of chemical conjugation and ADC are described in the prior art (e.g. WO 2009099741 A1, WO 2010009124, WO 2012073217 referenced herein), a compound like PNUEDA -Gly _n , or ADCs containing the sortase-conjugated PNU-EDA-Gly _n provided herein, have not yet been described in the prior art and are not the specific structure of a PNU derivative with an EDA spacer and a Glyn linker described or claimed in any of the prior art documents. Stable adducts in which PNU derivatives are stably linked to proteins through peptide bonds rather than ester bonds, and maleimide linkers may be superior in terms of stability and pharmacokinetic behavior in vivo due to the generally high stability of peptide bonds in serum, as described in the examples below. In addition, Glyn-segmented PNU derivatives that are expected to display stable drug conjugates after SMAC conjugation are shown in FIG. 6A and 6B.

Experiments and figures

While the invention has been illustrated and detailed in the drawings and in the foregoing description, such illustration and description are to be considered illustrative or exemplary and not restrictive; the invention is not limited to the described embodiments. Other modifications of the described embodiments are understood by those skilled in the art and can be made by them in the practice of the claimed invention after studying the drawings, the description, and the attached patent claims. In a patent claim, the word containing does not exclude other elements or steps, and the singular does not exclude plurality. The mere fact that certain measures are listed in mutually distinct dependent clauses does not indicate that a combination of these measures cannot be used in the interests of profit. Any references in the claims should not be construed as limiting the scope.

All amino acid sequences presented in this document are shown from the N-terminus to the C-terminus; all nucleic acid sequences presented in this document are shown 5'->3'.

Example 1 Generation of site-specific C-terminally conjugated monoclonal antibodies brentuximab and trastuzumab with PN-EDA-Glyn payload using sortase-mediated antibody conjugation (SMAC) technology.

The heavy and light chain variable region sequences of the monoclonal antibody brentuximab (clone cAc10) specific for the human CB30 target were obtained from US Pat. Kadcyla® extracted from it were obtained from the online IMGT database (VH: http://www.imgt.org/3DstructureDB/cgi/details.cgi?pdbcode=7637&Part=Chain&Chain=7637H & VL: http://www .imgt.org/3DstructureDB/cgi/details.cgi?pdbcode=7637&Part=Chain&Chain=7637L Chimeric mAb cAc10 and humanized mAb trastuzumab were generated, their heavy and light chains labeled at the C-terminus with a sortase A recognition sequence and an additional purification label Strep II affinity (HC-tagged sequence: LPETGGWSHPQFEK, LC-tagged sequence: GGGGSLPETGGWSFLPQFEK), s

- 14 040604 using methods known to those skilled in the art (see FIGS. 11A and 11B).

The anthracycline derivative PNU-EDA-Gly5 (FIG. 3A) was provided by Levana Biopharma, San Diego, Calif., which synthesized the pentaglycine peptide to the carbonyl group of PNU159682 using an ethylenediamine (EDA) linker according to the synthesis scheme depicted in FIG. 3b. To do this, commercially available PNU159682 was first oxidized to obtain its carboxylic acid (1 in Fig. 3B) with NaIO ₄ in 60% methanol at room temperature for 3 hours. Then to a solution of 1 (51 mg, 81 pmol) in 6 ml DCM N-hydroxysuccidimide (NHS, 46 mg, 400 pmol) and ethyl(dimethylaminopropyl)carbodiimide (EDC, 100 mg, 523 pmol) in dichloromethane (DCM) were added. After 30 min the mixture was washed with water (2x6 ml), dried over Na ₂ SO ₄ and evaporated. The residue was then dissolved in 2 ml dimethylformamide (DMF) before adding the amine (2 in Figure 3B, 55 mg, 81 pmol as trifluoroacetate salt), then N,N-diisopropylethylamine (DIEA, 50 pL) was added. The mixture was stirred for 1 h before the addition of piperidine (40 ml), after which it was stirred for another 20 min. The mixture was purified by HPLC to give PNU-EDA-Gly ₅ (3 in Figure 3B, 34 mg, 44%) as a red solid; MS m/z 955.2 (M+H).

PNU-EDA-Gly ₅ was conjugated to the mAb by incubating the LTETG-tagged mAb [10 pM] with PNUEDA-Gly ₅ , [200 pM] in the presence of 0.62 pm sortase A in 50 mM Hepes, 150 mM NaCl, 5 mM CaCl ₂ , pH 7.5, for 3.5 hours at 25°C. The reaction was stopped by passing it through a Protein A HiTrap (GE Healthcare) column equilibrated with 25 mM sodium phosphate, pH 7.5 followed by a 5 column volume (CV) wash with buffer. The bound conjugate was eluted with 5 CV elution buffer (0.1 M succinic acid, pH 2.8), 1 CV fractions were collected in tubes containing 25% v/v. 1 M Tris Base [Tris Base] to neutralize acid. Protein-containing fractions were pooled and added to 10 mM sodium succinate, pH 5.0, 100 mg/ml trehalose, 0.1% w/v. polysorbate or phosphate 20 by G25 column chromatography using NAP 25 columns (GE Healthcare) according to the manufacturer's instructions.

The total content of each conjugate was assessed by chromatography on a TOSOH TSKgel G3000SWXL 7.8 mm x 30 cm column, 5 rm, at a rate of 0.5 ml/min in 10% IPA, 0.2 M potassium phosphate, 0.25 M potassium chloride, pH 6.95. Drug loading was assessed by both hydrophobic interactive chromatography (HIC) and reverse phase chromatography. HIC was performed on a TOSOH Butyl-NPR 4.6 mm x 3.5 cm, 2.5 rm column at 0.8 ml/min with a 12-minute linear gradient between A-1.5M (NH ₄ ) ₂ SO ₄ , 25 mM NaPi, pH=6.95±0.05 and B-75% 25 mM NaPi, pH=6.95±0.05, 25% IPA. Reverse phase chromatography was performed on a Polymer Labs PLRP 2.1 mm x 5 cm, 5 rm column at 1 ml/min/80°C with a 25 minute linear gradient between 0.05% TFA/H2O and 0.04% TFA/CH3CN. Samples were first recovered by incubation with DTT at pH=8.0 at 37° C. for 15 min. Both PNU-EDA-Gly ₅ ADC conjugates were predominantly monomeric and had a drug-antibody ratio close to the theoretical maximum, respectively 4. 2 shows the results of obtaining ADC.

Table 2. Information on the obtained ADC based on PNU-EDA-Gly ₅ .

mAb target HC-label LC label % mono DAR Brentuximab CD30 Yes Yes 99.6 4.0 Trastuzumab HER-2 Yes Yes 98.2 3.9

HC - heavy chain; LC - light chain; % mono - % monomer content; DAR - drug-antibody ratio.

Example 2 In Vitro Cytotoxicity Assay Using Sortase-A Conjugated Brentuximab-PNU-EDA-Gly ₅ and Trastuzumab-PNU-EDA-Gly ₅ ADC Conjugates.

The cytotoxicity of brentuximab-PNU-EDA-Gly ₅ was tested using Karpas-299, a non-Hodgkin's lymphoma cell line expressing high levels of CD30, and L428, a Hodgkin's lymphoma cell line expressing low to moderate levels of CD30 (FIG. 4). As controls, the potency of cAc10-PNU-EDA-Gly ₅ was compared with that of the commercially available CD30-specific cAc10-vcPAB-MMAE AdCetris® conjugate (as a positive control) and the commercially available HER-2-specific trastuzumab-DM1 conjugate Kadcyla® (in as a negative control). For this, cells were seeded in 96-well plates in 100 ml RPMI/10% FCS at a density of 10 ⁴ cells per well and grown at 37°C in a humidified incubator under 5% CO2 atmosphere. After a 24-hour incubation, 25 cl of media was carefully removed from each well and replaced with 25 cl of 3.5-fold serial dilutions of each ADC in culture medium, resulting in final ADC concentrations ranging from 20 rg/ml to 0.25 ng/ml. Each dilution was performed in duplicate. After another 4 days, the plates were removed from the incubator and equilibrated to room temperature. Approximately 30 min later, 100 cl of CellTiter-Glo® Luminescent Solution (Promega, Cat. No. G7570) was added to each well and after shaking the plates at 450 rpm for 5 min.

- 15 040604 and subsequent incubation for 10 min without shaking, the luminescence was measured on Tecan Infinity F200, the integration time was 1 s per well.

As expected, the anti-CD30 ADC Adcetris® used as a positive control was effective in killing CD30 ^HI Karpas-299 cells at an EC50 of 8.2 ng/mL (Fig. 4A), but was ineffective in killing CD30 ^LO L428 cells (Fig. 4B). In contrast, the anti-HER-2 ADC Kadcyla® used as a negative control showed no specific cell killing and was ineffective on any of the cell lines (FIG. 4). Notably, the sortase-conjugated cAc10PNU-EDA-Gly5 ADC was effective in killing Karpas-299 CD30 ^HI cells at an EC50 value of 6.9 ng/mL (Fig. 4A). cAc10-PNU-EDA-Gly5 only killed CD30 ^HI L428 cells at higher concentrations, similar to those used by control ADCs, indicating that the efficacy of this ADC is indeed specific and mediated by binding to CD30 (Fig. 4B). Thus, sortase-mediated conjugation of PNU-EDA-Gly ₅ produced an ADC conjugate with very high activity, even exceeding that of the Adcetris® ADC.

The tumor cell killing activity of SMAC-generated ADC trastuzumabPNU-EDA-Gly ₅ was investigated using SKBR3 cells, a human breast cancer cell line overexpressing HER-2, and T47D cells, a breast cancer cell line naturally expressing low levels of HER -2 and compared with the commercially available HER-2 specific ADC trastuzumab-DM1 Kadcyla® conjugate (FIG. 5). For this, cells were seeded in 96-well plates in 100 μl DMEM/10% FCS at a density of 10 ⁴ cells per well, analyzes were performed exactly as described above.

As expected, the Kadcyla® positive ADC control was effective in killing HER-2 overexpressing human breast cancer cells SKBR3 at an EC50 of 23.7 ng/mL (FIG. 5A), but was ineffective in killing HER-2 ^LO T47D cells (FIG. 5B). Notably, trastuzumab-PNU-EDAGly5, generated using SMAC technology, showed excellent cytotoxicity and not only killed HER-2 overexpressing SKBR3 cells, but also HER2 ^LO T47D cells, at EC50 values of 4.8 and 11.0 ng, respectively. /ml (Fig. 5). Thus, sortase-mediated conjugation of PNUEDA-Gly5 to trastuzumab produces an ADC with very high potency, superior to that of the commercially available and FDA-approved reference ADC Kadcyla®, and even effective on HER2 ^LO human breast cancer cells.

Example 3 Serum resistance in vitro of sortase A-conjugated cAc10-PNU-EDA-Gly ₅ ADC compared to a maleimide linker containing trastuzumab emtansine (Kadcyla®).

In vitro serum resistance of brentuximab-PNU-EDA-Gly ₅ (cAc10-PNU-EDA-Gly ₅ ) and Kadcyla ADC conjugates was assessed using an ELISA-based serum resistance assay. Briefly, cAc10-PNU-EDA-Gly ₅ was diluted in mouse (Sigma, M5905), rat (Sigma, R9759) and human (Sigma, H6914) serum and incubated at 37°C. Samples were flash frozen in liquid nitrogen on days 0, 3, 7, 14 and stored at minus 80° C. until ELISA analysis. For rodent sera, dilution series of cAc10-PNU-EDA-Gly ₅ serum samples were fixed on ELISA plates coated with 2 cg/ml mouse anti-PNU mAb (in-house production by immunizing mice with human IgG-PNU conjugate and screening with BSA-PNU conjugate) to bind to ADC, or coated with anti-human Fc F(ab') 2 (Jackson Immunoresearch) to bind to total IgG, and detected with a 1:2500 dilution of HRP-conjugated anti-human IgG F(ab') 2 (Jackson Immunoresearch). For primate sera, 2 cg/ml recombinant human CD30 (Sino Biologicals, 10777-H08H) was applied to ELISA plates and a 1:2500 dilution of HRP-conjugated anti-human IgG F(ab')2 (Jackson Immunoresearch) or 1 ng/ml mouse anti- PNU IgG (in-house) and then HRP-conjugated anti-mouse Fc F(ab')2 (Jackson Immunoresearch) were used to detect total IgG and ADC, respectively. In the case of Kadcyla, the same protocol as above was used to determine resistance in mouse, rat and human sera, but using a proprietary anti-maytansine mAb to bind to ADC. Serum ADC and total IgG concentrations were calculated from half of the maximum sample titrations by comparison with a sample of the same ADC of known concentration.

In FIG. 7A shows the excellent stability of cAc10-PNU-EDA-Gly ₅ ADC, in particular compared to the Kadcyla-containing maleimide linker (FIG. 7B), with virtually no reduction in ADC levels throughout the experiment in any of the four sera tested. By fitting the time points between days 0 and 14 to a single-phase exponential decay function limited to reach a final concentration of 0, the half-lives of cAc10-PNUEDA-Gly ₅ and Kadcyla were determined in each serum. The half-life of Kadcyla was 3.7 days, 4.4 days, and 2.9 days in mouse, rat, and human sera, respectively, while the half-life of cAc10-PNU-EDA-Gly ₅ was over 14 days in mouse, rat, and human sera. .

Example 4 In vivo stability of sortase A-conjugated Ac10-Gly ₅ -PNU in mice.

ADC AcC-Gly ₅ -PNU was thawed at room temperature and diluted to 0.2 mg/ml in sterile

- 16 040604

PBS up to a dosing concentration of 1 mg/kg. Samples were injected intravenously at a volume of 5 ml/kg to nine female Swiss Webster mice. Blood was collected from animals after 1 h, 24 h, 72 h, 7 days, 14 days and 21 days. Individual animals in accordance with ethical standards were used only for two time points of blood sampling, at least one week apart. Thus, three mice were bled at 1 hour and 7 days, three other mice were bled at 24 hours and 14 days, and another three mice were bled at 72 hours and 21 days, for a total of nine mice per group. For each group of animals, approximately 200 μl of blood was collected by lancet puncture of the submandibular vein during the first collection and approximately 600 μl of blood by lancet puncture of the submandibular vein during the final collection (final phlebotomy). All blood was collected in tubes containing K2-EDTA. Plasma was isolated from blood by centrifugation at 1500g for 10 min and transferred to sterile cryo-vials for storage at -80°C until analysis by ELISA as described in Example 4.

The data in FIG. 8 show the high stability of the ADC created using SMAC technology. Throughout the experiment, ADC concentrations were only slightly lower than those measured for total IgG, implying that the drug-antibody linker is stable in vivo. By fitting the time points between days 3 and 21 to a single phase exponential decay function limited to reach a final concentration of 0, the slow phase in vivo half-life was determined to be 8.3 and 7.8 days for total IgG and ADC, respectively.

Example 5 Description and characterization of EMT-6 clones expressing HER-2.

Cytotoxicity of anti-HER-2 ADCs was studied using the EMT-6 mouse breast tumor cell line engineered to overexpress human HER-2. EMT-6 cells were cultured as monolayers in DMEM (Dulbecco's Modified Eagle Medium - High Glucose) supplemented with 10% (v/v) FCS (fetal calf serum), 1% (v/v) 10,000 IU/mL penicillin-streptomycin and 1% (v/v) 200 mM L-glutamine.

EMT-6 cells were electroporated with an expression vector encoding the human HER-2 gene and a puromycin resistance marker, and cell pools stably expressing human HER-2 were selected using methods known to those skilled in the art.

Expression of HER-2 was confirmed by flow cytometry. Briefly, after trypsinization, 106 cells were centrifuged in FACS tubes; the resulting beads were resuspended in PBS (phosphate buffered saline) supplemented with 2% FCS. Cells were then incubated with antiHER-2 antibody trastuzumab (30 min, 4°C) followed by centrifugation and washing (3 ml PBS with 2% FCS). The cells were then resuspended as before and incubated with anti-human IgG antibody (Fc-gamma specific) PE (Ebioscience) in the dark (30 min, 4°C), before washing (4 ml PBS with 2% FCS). Then flow cytometry was performed on a FACS Calibur (BD).

HER-2 transfected EMT-6 cells were single cell sorted by flow cytometry using FACS ARIA II to isolate single cell clones. They were expanded and HER-2 expression was confirmed by flow cytometry.

In FIG. 9 shows FACS analysis data for a clone selected for in vivo studies (Example 6).

Example 6 In vivo performance of sortase A-conjugated ADC trastuzumab-PNU-EDA-Gl·y ₅ in an orthotopic model of breast cancer.

The in vivo efficacy of trastuzumab-PNU-EDA-Gl·y ₅ was evaluated in an immunocompetent orthotopic mouse model of HER-2 positive breast cancer. For this, 106 EMT6 mouse breast cancer cells expressing human HER-2 (Example 6), previously identified as suitable for in vivo growth, were implanted in the right mammary fat pads of female Balb/c mice. In addition, HER-2 negative EMT6 cells were implanted in control animals. Next, the primary tumor volumes were measured with a caliper. After 13 days, when an average tumor volume of 100-150 mm ³ was reached, animals with tumors were randomized into groups of 6 animals depending on tumor size. Animals were treated on the same day (Day 13, i.e. randomization day) and 7 days later (Day 20) by intravenous injection of Kadcyla® reference ADC (15 mg/kg), trastuzumab-PNU-EDA-Gl y ₅ (1 mg/kg) or vehicle control [vehicle control]. The size of the tumors was controlled by measuring with a caliper and in animals whose tumor volume reached 1000-1500 mm ³ the test was terminated (FIG. 10).

Tumors in vehicle control mice grew rapidly and reached a mean size of approximately 1000 mm ³ within 30 days of cell transplantation (FIG. 10A). Treatment with Kadcyla® had little effect on tumor growth in most animals. Only one of the six animals showed a significant delay in tumor growth (Fig. 10C). In striking contrast, in all animals treated with trastuzumab-PNU-EDA-Gly5, tumors continuously regressed during treatment and were essentially undetectable by day 30 post-cell transplantation (FIG. 10D). In most animals no tumor was detected until day 60, and tumor recurrence was observed in only one animal around day 40. Significantly, the antitumor activity of trastuzumab-PNU

- 17 040604

EDA-Gly ₅ was highly specific and treatment of mice bearing HER-2 negative tumors did not result in tumor regression (FIG. 10B). Taken together, the data demonstrates that trastuzumab-EDA-Gly ₅ -PNU site-specifically conjugated ADCs produced ADCs with in vivo tumor cell killing activity far superior to the reference Kadcyla® ADC.

List of figures

Fig. 1. Schematic representation of site-specific sortase-mediated antibody conjugation (SMAC technology). Monoclonal antibodies should be generated with C-terminal LPXTG sortase tags. It is necessary to create a toxic payload to contain an oligoglycine peptide stretch (Glyn-cut) with a certain number of glycine residues in a row (n>1 and <21, preferably n>3 and <10, preferably n=3 or n=5, most preferably n =5). The sortase A enzyme from Staphylococcus aureus specifically recognizes the LPXTG pentapeptide motif and catalyzes the transpeptidation of the oligoglycine peptide segment to the threonine-glycine-LPXTG peptide bond, thereby creating a new stable peptide bond between the threonine and the N-terminal glycine of the oligoglycine segment.

Fig. 2. Structure of PNU-159682 as described in the prior art (eg WO 2009099741 or Quintieri et al. (2005)), including the official anthracycline numbering system for the reactive carbons of the aglycone tetracyclic structure.

Fig. 3(A). The structure of PNU derivative-EDA-Gl·y ₅ , herein referred to as PNU-EDAGly ₅ , as used for SMAC conjugation with C-terminal LPETG sortase-labeled monoclonal antibodies using the sortase enzyme as described in the examples herein ; (B) Scheme for the synthesis of the anthracycline derivative PNU-EDA-Gly ₅ .

Fig. 4. Dose-response of cytotoxic effects of these ADCs on the Karpas-299 human non-Hodgkin's lymphoma cell line expressing high levels of target CD30 on the cell surface (A) and on L428 cells of the human Hodgkin's lymphoma cell line expressing very low levels of target CD30 on the cell surface (IN). Adcetris refers to the commercially available anti-CD30 ADC brentuximab-vedotin. Kadcyla refers to the commercially available anti-HER-2/neu ADC T-DM1 (trastuzumab-emtansine). Both cell lines are negative for HER-2/neu and therefore Kadcyla acts as a negative ADC control that does not have to perform cell killing in a target-specific manner. Cells were incubated with serial dilutions of ADC for 4 days, after which CellTiter-Glo® (Promega) luminescent solution was added and viable cells were quantified by measuring luminescence on a Tecan Infinity F200.

Fig. 5. Dose-response of cytotoxic effects of these ADCs on the human breast cancer cell line SKBR3 expressing high levels of HER-2/neu (A) and the human breast cancer cell line T47D expressing low levels of HER-2/neu (B). Cells were incubated with serial dilutions of ADC for 4 days, after which CellTiter-Glo® (Promega) luminescent solution was added and viable cells were quantified by measuring luminescence on a Tecan Infinity F200.

Fig. 6. Additional PNU-159682-linked anthracycline derivatives useful for site-specific conjugation to LPXTG-labeled binding proteins or antibodies using SMAC technology to produce BPDCs or ADCs. Only preferred versions with Gl·y ₅ cut are shown. In FIG. 6A shows a derivative in which the Gly ₅ amino acid stretch is directly connected via its carboxy terminus to the A-ring of the tetracyclic aglycone structure of the PNU derivative. In FIG. 6B depicts a derivative in which the preferred ethyleneamine linker and Gly ₅ amino acid stretch are directly connected to the A-ring of the tetracyclic aglycone structure of the PNU derivative.

Fig. 7 (A) Measurement of in vitro concentration of brentuximab-PNU-EDA-Gl y ₅ ADC (designated as cAc10-PNU ADC) and total IgG in mouse (A), rat (B), human (C) serum for 14 days . (B) In vitro measurement of trastuzumab-emtansine (Kadcyla®) ADC concentration and total IgG in mouse (A), rat (B), and human (C) serum over 14 days.

Fig. 8. In vivo plasma concentrations of ADC and total IgG measured at 6 time points over a 21 day period following Ac10-Gly ₅ -PNU ADC administration in mice.

Fig. 9. FACS analysis data of HER-2 clone EMT-6 selected for in vivo studies after incubation with anti-HER-2 antibody trastuzumab and then incubation with fluorophore-containing anti-human IgG antibody (Fc-gamma-specific) PE .

Fig. 10. In vivo evaluation of HER-2-specific ADCs in an orthotopic immunocompetent mouse model of HER2-positive breast cancer. EMT6 mouse breast cancer cells expressing human HER-2 (A, C, D) or the irrelevant ROR-1 antigen were grown in the milk fat pads of Balb/c mice. On days 13 and 20, animals were treated intravenously with vehicle control (A), 1 mg/kg trastuzumab-PNU159682 (B, D) or 15 mg/kg Kadcyla (C). Tumor growth was controlled until the animals had to be sacrificed for ethical reasons.

- 18 040604

Fig. 11 A and B: Amino acid compositions of C-terminal SMAC™ conjugated IgH and IgL chains of the PNU toxin derivative trastuzumab (A) and brentuximab (B) containing the ADCs used for research containing the PNU derivative depicted in FIG. 3B linked through the amino group of the Gly ₅ -segment to the 4th amino acid of the sortase tag (in bold) via a peptide bond after conjugation of the sortase enzyme.

Bibliography

Beerley et al. (2015) PloS One 10 eO13U77

Dorr et al. (2014) PNAS 111, 13343-8

Quintieri L et al (2005), Clin Cancer Res. 2005 Feb 15;11(4): 1608-17.

Roguska et al. (1994) PNAS 91, pp969-971

Perez et al. (2014) Drug Discovery Today 19, pp.869-881

Alley et al. (2008) Bioconjug. Chern. 19, pp. 759-765

Panowski et al. (2014) mAbs 6, pp. 34-45

Wolff AC et al (2013). J Clin Oncol. 2013 Nov 1;31(31):3997-4013

Young KH (2014) Clinical Advances in Hematology & Oncology, Volume 12, Issue 4,

Supplement 10

Spirig et al. (2011) Mol. microbiol. 82, 1044-1059

Ducry & Stump (2010) Bioconjug. Chern. 21.5-13,

McCombs et al. (2015) The AAPS Journal 17, 339-351

Mullard (2013) Nature Rev Drug Disc 12, 329-332

Doctor et al. (2014) Mol Cancer Ther 13, 2618-2629

Hartley & Hochhauser (2012) Curr. Opin. Pharmacol. 12, 398-402

Minotti (2004) Pharmacol. Rev 56, 185-229

Cancer and Chemotherpay - Antineoplastic Agents Vol. Ill, Stanley T. Crooke and Archie W.

Prestayko (eds.), Academic Press 1981).

Claims (31)

CLAIM 1. An anthracycline derivative (PNU) conjugate containing an anthracycline derivative PNU159682 having the following formula (i): o he o formula (i) moreover, the indicated conjugate contains on its wavy line ? linker structure L1-L2-L3, where L1 is an alkylenediamine linker? L2 is an oligoglycine peptide linker (Glyn), L ₃ represents amino acid residues 1-4 of the pentapeptide sortase enzyme recognition motif lacking a C-terminal glycine residue, and two of L1-L ₃ are mandatory, and L2 is connected to said anthracycline derivative of formula (i) using an L1 linker, where linker L1 is connected to said anthracycline derivative via a first amide bond, while it is linked to the carboxy-terminus of linker L ₂ via a second amide bond, wherein said L1 linker-L2 linker conjugate has the following formula: -NH-(CH2)m-NH-(Gly) _n -NH ₂ (formula (v)), - 19 040604 where the wavy line shows a compound with said anthracycline derivative of formula (i), where m is an integer not less than 1 and not more than 11, and n is an integer not less than 1 and not more than 21. 2. An anthracycline derivative (PNU) conjugate containing an anthracycline derivative having the following formula (ii): moreover, the specified conjugate contains on its wavy line the linker structure L1-L2-L3, where L1 is an alkyleneamine linker, L2 is an oligoglycine peptide linker (Glyn), L3 represents amino acid residues 1-4 of the pentapeptide recognition motif of the sortase enzyme lacking a C-terminal glycine residue, and two of L1-L3 are mandatory, and L2 is connected to said anthracycline derivative of formula (ii) using an L1 linker, where the L1 linker connected to the carboxy-terminus of the L2 linker via an amide bond, said L1 linker-L2 linker conjugate having the following formula: ^-(CH ₂ ) _m -NH-(Gly)n-NH2 (formula (vi)), where the wavy line shows the connection with the specified anthracycline derivative of the formula (ii), where m is an integer not less than 1 and not more than 11 , and n is an integer greater than or equal to 1 and greater than or equal to 21. 3. An anthracycline derivative (PNU) conjugate according to claim 1, wherein L3 is selected from TXPL, SXPL, and TXAL, where X is any of the 20 peptidogenic amino acids. 4. Protein-Drug Binding Conjugate (BPDC) having the following formula: formula (Hi), where L1 is an alkylenediamine linker, L2 is an oligoglycine peptide linker (Glyn) and L3 is amino acid residues 1-4 of the pentapeptide sortase enzyme recognition motif lacking a C-terminal glycine residue, where two of L1-L3 are mandatory and where linker X is absent and linker Y absent or composed of at least one and at least 20 amino acids between the C-terminal amino acid residue of the binding protein and the sortase enzyme recognition motif, L3, BP is a binding protein, n is an integer between >1 and <10 and where the L2 linker is connected to the anthracycline derivative of formula (iii) via the L1 linker, where the L1 linker is connected to the anthracycline derivative of formula (iii) via the first amide bond, while linker L1 is connected to the carboxy-terminus of linker L2 by a second amide bond, where said conjugate of linker L1 and linker L2 has the formula (v) j NH-(CH ₂ )m-NH-(Gly) _n -NH2 (formula ( ν)), where the wavy line f shows a compound with the indicated anthracycline derivative of formula (v), where m is an integer not less than 1 and not more than 11, and n is an integer - 20 040604 lo not less than 1 and not more than 21. 5. Protein-Drug Binding Conjugate (BPDC) having the following formula: ΰ he Civ), where L ₁ is an alkyleneamine linker, L2 is an oligoglycine peptide linker (Glyn) and L3 is amino acid residues 1-4 of the pentapeptide sortase enzyme recognition motif lacking a C-terminal glycine residue, where two of L1-L3 are mandatory and where linker X is absent and linker Y absent or composed of at least one and at least 20 amino acids between the C-terminal amino acid residue of the binding protein and the sortase enzyme recognition motif, L3, BP is a binding protein, n is an integer not less than 1 and not more than 10, and where the L2 linker is connected to the anthracycline derivative of formula (iv) via an L1 linker, where the L1 linker is connected to the carboxy-terminus of the L2 linker via an amide bond, where said conjugate of linker L1 and linker L2 has the formula (vi) j (CH ₂ ) _m -NH-(Gly) _n -NH2 (formula (vi)) where m is an integer not less than 1 and not more than 11, and n is an integer not less than 1 and not more than 21. 6. A protein-drug-binding conjugate (BPDC) according to claim 4, wherein L3 is selected from TXPL, SXPL, and TXAL, where X is any of the 20 peptiogenic amino acids. 7. A protein-drug-binding conjugate (BPDC) according to claim 4, wherein L3 is directly connected to the C-terminus of the binding protein. 8. A protein-drug-binding conjugate (BPDC) according to claim 4, characterized in that L3 is connected to the C-terminus of the binding protein. 9. A protein-drug-binding conjugate (BPDC) according to claim 4, wherein the Y linker is any amino acid chain of up to 20 amino acids that provides optimal conjugation of the binding protein to L3. 10. A protein-drug-binding conjugate (BPDC) according to claim 9, wherein the Y linker is a 5 amino acid linker located between the C-terminus of the immunoglobulin light chain and L3. 11. A protein-drug-binding conjugate (BPDC) according to claim 10, wherein the Y linker has the amino acid sequence GGGGS. 12. A protein-drug-binding conjugate (BPDC) according to any one of claims 4 and 6-11, wherein the linker structure contains a peptide motif that results from specific cleavage of the sortase enzyme recognition motif. 13. A protein-drug-binding conjugate (BPDC) according to claim 12, wherein said sortase enzyme recognition motif contains a pentapeptide. 14. A protein-drug-binding conjugate (BPDC) according to claim 12 or 13, characterized in that said sortase enzyme recognition motif contains at least one of the following amino acid sequences: LPXTG, LPXSG and/or LAXTG. 15. A protein-drug-binding conjugate (BPDC) according to any one of claims 4 and 6-14, wherein the anthracycline derivative (PNU) is conjugated by one or more linkers to the carboxy terminus of the binding protein. 16. A protein-drug-binding conjugate (BPDC) according to any one of claims 4 and 6-15, wherein the binding protein is conjugated to the free amino terminus of the oligoglycine peptide (Gly _n ) via an amide bond. 17. A protein-drug-binding conjugate (BPDC) according to any one of claims 4 and 6-16, wherein the binding protein is at least one selected from the group consisting of an antibody, a modified antibody format, a derivative, or an antibody fragment, an antibody-based binding protein, an oligopeptide binder, and/or an anti-antibody mimetic - 21 040604 bodies. 18. A protein-drug-binding conjugate (BPDC) according to any one of claims 4 and 6-17, wherein the binding protein binds to at least one entity selected from the group consisting of receptor, antigen, growth factor, cytokine and/or hormone. 19. A protein-drug-binding conjugate (BPDC) according to any one of claims 4 and 6-18, wherein the binding protein has at least two subunits. 20. A protein-drug-binding conjugate (BPDC) according to claim 19, wherein at least one subunit contains the anthracycline derivative PNU-159682 according to claim 1. 21. A protein-drug-binding conjugate (BPDC) according to any one of claims 4 and 6-20, wherein the binding protein binds to HER-2. 22. A protein-drug-binding conjugate (BPDC) according to any one of claims 4 and 6-21, wherein the binding protein is an antibody that binds to HER-2. 23. A protein-drug-binding conjugate (BPDC) according to any one of claims 4 and 6-22, wherein the antibody a) contains CDR regions 1-6 of trastuzumab, b) contains the heavy chain variable domain and the light chain variable domain of trastuzumab, c) has an amino acid sequence identity of 90% or greater with regions or domains from a) or b), d) is trastuzumab or a target-binding fragment or derivative thereof and/or e) competes with trastuzumab for binding to Her-2. 24. A protein-drug-binding conjugate (BPDC) according to any one of claims 4 and 6-20, wherein the binding protein binds to CD30. 25. A protein-drug-binding conjugate (BPDC) according to any one of claims 4 and 6-20 or 23, wherein the binding protein is an antibody that binds to CD30. 26. A protein-drug-binding conjugate (BPDC) according to any one of claims 4 and 6-20 and 2425, wherein the antibody a) contains the CDR region 1-6 of brentuximab, b) contains the heavy chain variable domain and the light chain variable domain of brentuximab, c) has an amino acid sequence identity of 90% or greater with regions or domains from a) or b), d) is brentuximab or a target-binding fragment or derivative thereof and/or e) competes with brentuximab for binding to CD30. 27. A method for producing a protein-drug-binding conjugate (BPDC) according to any one of claims 4 and 6-26, characterized in that the binding protein containing the sortase enzyme recognition motif is conjugated by the sortase enzyme to at least one anthracycline derivative conjugate 1, which contains an oligoglycine peptide (Glyn). 28. The use of a protein-drug-binding conjugate (BPDC) according to any one of claims 4 and 6-26 for the treatment of a human or animal suffering from, at risk of developing and/or diagnosed with a neoplastic disease. 29. Use according to claim 28, characterized in that the neoplastic disease is a cancer that has a HER-2 expression index of 1+, 2+, or 3+, as determined by immunohistochemistry or in situ hybridization; a cancer that is CD30 positive, as determined by immunohistochemistry, ELISA, or flow cytometry. 30. Use according to claim 29, wherein the cancer is breast cancer. 31. Use according to claim 29, wherein the cancer is at least one of lymphoma, non-Hodgkin's lymphoma, systemic anaplastic large cell lymphoma (SACL).