CHIMER CONTAINING A TARGETING PORTION LINKED TO A SCISSION-ACTIVATED DUOCARMYCIN-TYPE PRODRUG

DESCRIPTION

CROSS-REFERENCE TO RELATED APPLICATIONS

This application claims priority of provisional applications Serial No. 61/002,978 and Serial No. 60/987,647, both of which were filed on November 13, 2007, and whose disclosures are incorporated by reference.

GOVERNMENTAL SUPPORT The present invention was made with governmental support pursuant to USPHS grant CA41986 from the National Institutes of Health. The government has certain rights in the invention.

TECHNICAL FIELD

The present invention relates to a scissile chimeric molecule that contains a targeting polypeptide portion covalently attached to a cytotoxic hydroxy- functional prodrug pharmaceutical portion and permits controlled release of the hydroxy- functional prodrug, as well as compositions and methods of using the same. In a particularly preferred embodiment, the invention relates to a targeting polypeptide covalently bonded by a linker group to a scission-activated duocarmycin-type prodrug .

BACKGROUND ART

(+)-CC-1065, the duocarmycins, and yatakemycin (below) constitute exceptionally potent naturally

occurring antitumor agents that derive their biological properties through a characteristic sequence-selective DNA alkylation reaction [Chidester et al., J. Am. Chem. Soc. 1981, 203;7629; Takahashi et al., J. Antibiot. 1988, 42:1915; Ichimura et al . , J. Antibiot. 1990, 43:1037; Yasuzawa et al . , Chem. Pharm. Bull. 1995, 43:378; Igarashi et al . , J. Antibiot. 2003, 56:107; Structure revision: Tichenor et al., J. Am. Chem. Soc. 2004, 226:8396; Warpehoski et al., Chem. Res. Toxicol. 1988, 2:315; Hurley et al., Ace. Chem. Res. 1986, 29:230; Boger et al . , Bioorg. Med. Chem. 1994, 2:115; Boger et al . , J. Am. Chem. Soc. 1990, 222:4623; Boger et al . , J. Am. Chem. Soc. 1990, 222:8961; Boger et al . , J. Am. Chem. Soc.

1991, 123:6645; Boger et al . , Bioorg. Med. Chem. Lett. 1992, 2:759; Duocarmycin SA: Boger et al . J. Am. Chem. Soc. 1994, 116:1635; Parrish et al . , J. Am. Chem. Soc. 2003, 125:10971; Tichenor et al . , J. Am. Chem. Soc. 2006, 128:15683; Trzupek et al . , Nature Chem. Biol. 2006, 2:79] .

Examination of the natural products, their synthetic unnatural enantiomers, their derivatives, and synthetic analogues has defined fundamental features that control the alkylation selectivity, impact the alkylation efficiency, and are responsible for DNA alkylation catalysis providing a detailed understanding of the relationships between structure, reactivity, and biological activity [Warpehoski et al., Chem. Res. Toxicol. 1988, 1:315; Hurley etal . , Ace. Chem. Res. 1986, 19:230; Boger et al . , Ace. Chem. Res. 1999, 32:1043; Boger et al . , M. Bioorg. Med. Chem. 1997, 5:263; Boger et al . , Angew. Chem., Int. Ed. Engl. 1996, 35:1438; Boger, Ace. Chem. Res. 1995, 28:20; Boger et al . , Proc. Natl. Acad. Sci . , U.S.A. 1995, 92:3642; Boger, Chemtracts: Org. Chem. 1991, 4:329; Review of synthetic studies: Boger et al., Chem Rev. 1997, 97:787. Tse et al . , Chem. Biol. 2004, 11:1607] .

One of the most important and widely explored class of analogues is CBI (1,2,9,9a- tetrahydrocyclopropa [c] benz [e] indol-4-one) [Boger et al., J. Am. Chem. Soc. 1989, 111:6461; Boger et al . J. Org. Chem. 1990, 55:5823], being synthetically [Boger et al . J. Am. Chem. Soc. 1989, 111:6461; Boger et al., J. Org. Chem. 1990, 55:5823; Boger et al . , J. Org. Chem. 1992, 57:2873; Boger et al . , J. Org. Chem. 1995, 60:1271; Drost et al . , J. Org. Chem. 1991, 56:2240; Aristoff et al . , J. Org. Chem. 1992,

57:6234; Mohamadi et al . J. Med. Chem. 1994, 37:232; Ling et al . Heterocyclic Commun. 1997, 3:405; Boger et al., Synlett 1997, 515; Boger et al . , Tetrahedron Lett. 1998, 39:2227; Kastrinsky et al . , J. Org. Chem. 2004, 69, 2284] more accessible than the natural products, yet indistinguishable in their DNA alkylation selectivity [Boger et al . , J. Am. Chem. Soc. 1992, 114:5487] . Moreover, the CBI derivatives proved to be four times more stable and, correspondingly, four times more potent than derivatives bearing the CC- 1065 alkylation subunit (7-MeCPI) approaching the stability and potency of duocarmycin SA and yatakemycin derivatives, and they exhibit efficacious in vivo antitumor activity in animal models at doses that reflect this potency [Boger et al . Bioorg. Med. Chem. Lett. 1991, 2:115; Boger et al . Bioorg. Med. Chem. 1995, 3:1429] .

Consequently, CBI and its derivatives have been the focus of much development as well as the prototype analogues on which new design concepts have been explored, developed, or introduced [Boger et al., J. Am. Chem. Soc. 1989, 111:6461; Boger et al . , J. Org. Chem. 1990, 55:5823; Boger et al . , J. Org. Chem. 1992, 57:2873; Boger et al . , J. Org. Chem. 1995, 60, 1271; Drost et al . , J. Org. Chem. 1991, 56:2240; Aristoff et al . , J. Org. Chem. 1992, 57:6234; Mohamadi et al . J. Med. Chem. 1994, 37:232; Ling et al . , Heterocyclic Commun. 1997, 3:405; Boger et al., Synlett 1997, 515; Boger et al . Tetrahedron Lett. 1998, 39:2227; Kastrinsky et al . , J. Org. Chem. 2004, 69:2284; Boger et al . , J. Am. Chem. Soc. 1992, 114:5487; Boger et al . , Bioorg. Med. Chem. Lett. 1991, 2:115; Boger et al . , Bioorg. Med. Chem. 1995, 3:1429; Boger et al . , J. Am. Chem. Soc. 1994,

116:7996; Boger et al . , Tetrahedron Lett. 1990, 31:193; Boger et al . , Bioorg. Med. Chem. Lett. 1991, 1:55; Boger et al . , J. Am. Chem. Soc. 1991, 113:2119; Boger et al . , J. Am. Chem. Soc. 1994, 116:5523; Boger et al., Bioorg. Med. Chem. 1995, 3:611; Boger et al . , Bioorg. Med. Chem. 1995, 3:761; Boger et al . , Bioorg. Med. Chem. 1997, 5:233; Boger et al . , J. Org. Chem. 1997, 62:8875; Boger et al . , J. Am. Chem. Soc. 1998, 120:11554; Boger et al . , J. Org. Chem. 1999, 64:5666; Boger et al . , J. Org. Chem. 1999, 64:5241; Boger et al., J. Org. Chem. 2001, 66:2207; Boger et al . , J. Org. Chem. 2001, 66:5163; Boger et al . , J. Org. Chem. 2001, 66:6654; Boger et al . , Bioorg. Med. Chem. Lett. 2001, 11:2021; Parrish et al . , Bioorg. Med. Chem.

2003, 11:3815; Parrish et al . , Bioorg. Med. Chem.

2004, 12:5845; Parrish et al . , J. Am. Chem. Soc. 2004, 126:80. MCBI: Boger et al . , J. Org. Chem. 1996, 61:1710. CCBI: Boger et al . , J. Org. Chem. 1996, 61:4894; Boger et al . , Bioorg. Med. Chem. Lett.

1996, 6:659. CNA: Boger et al . , . J. Org. Chem. 1997, 62:5844. Iso-CI/Iso-CBI: Boger et al . , J. Org. Chem.

1997, 62:8875. CBIn: Boger et al . , J. Org. Chem.

1998, 63:8004. CPyI: Boger et al . , J. Org. Chem. 2000, 65:4088. CBA: Parrish et al . , J. Org. Chem. 2003, 6:8984. Parrish et al . , Org. Lett. 2003, 5:2577; Kumar et al . , Org. Biomol. Chem. 2003, 1:2630; Kumar et al . , Heterocyclic Commun. 2002, 8:521; Kumar et al . , Org. Lett. 2002, 4:1851; Jia et al., Bioorg. Med. Chem. 2000, 8:1607; Jia et al . , Synlett 2000, 5:603; Jia et al . Heterocyclic Commun.

1999, 5:497; Jia et al . , Heterocyclic Commun. 1998, 4:557; Kumar et al . , Lett. Org. Chem. 2004, 1:154; Philips et al . , MoI. Pharmacol. 2005, 67:877; Wang et al., J. Biol. Chem. 2002, 277:42431; Chang et al . , J.

Am. Chem. Soc. 2000, 122:4856; Minoshima et al . , J. Am. Chem. Soc. 2007, 129:5384; Shinohara, K.; et al . Cancer Sci. 2006, 27, 219; Bando et al . , Bioconjugate Chem. 2006, 17:715; Bando et al . , J. Am. Chem. Soc. 2005, 127:13890; Bando et al . , J. Am. Chem. Soc. 2004, 125:8948; Wang et al . , Bioorg. Med. Chem. 2003, 11:1569; Wang et al . J. Med. Chem. 2003, 45:634; Wang et al., J. Med. Chem. 2000, 43:1541; Wang et al . , Bioorg. Med. Chem. 2006, 14:7854; Chari et al . , Cancer Res. 1995, 55:4079; Lillo et al . , Chem. Biol. 2004, 11:897; Tietze et al . , Eur. J. Org. Chem. 2002, 10:1634; Tietze et al . , Angew. Chem. Int. Ed. 2002, 41:759; Tietze et al . , ChemBioChem 2001, 2:758; Tietze et al . , Angew. Chem. Int. Ed. 2006, 45:6574; Wang et al . , Bioorg. Med. Chem. 2003, 11:1569; Jeffrey et al . , J. Med. Chem. 2005, 48:1344; Kline et al., MoI. Pharmaceut. 2004, 1:9; Hay et al . , J. Med. Chem. 2003, 45:5533; Tercel et al . , J. Med. Chem. 2003, 45:2132; Gieseg et al . , Anti-Cancer Drug Design 1999, 14:77; Hay et al . , Bioorg. Med. Chem. Lett. 1999, 9:2237; Atwell et al . , J. Med. Chem. 1999, 42:3400; Atwell et al . , J. Org. Chem. 1998, 53:9414; Atwell et al . , Bioorg. Med. Chem. Lett. 1997, 7:1493; Townes et al., Med. Chem. Res. 2002, 11:248; Boger et al., J. Org. Chem. 1999, 59:8350] .

A unique feature of this class of molecules including the natural products themselves is the observation that synthetic phenol precursors (e.g., Compound 1) to the final products, entailing a Winstein Ar-3' spirocyclization with displacement of an appropriate leaving group, exhibit biological properties typically indistinguishable from the cyclopropane-containing final products (DNA alkylation rate or efficiency, in vitro cytotoxic

activity, and in vivo antitumor activity) . This dependable behavior of the precursor phenols has provided the basis on which the development of useful, stable, or safe prodrugs has been conducted [(a) Carzelesin: Aristoff, Adv. Med. Chem. 1993, 2:67; (b) KW-2189: Kobayashi et al . Cancer Res. 1994, 54:2404; Amishiro et al . , Bioorg. Med. Chem. 2000, 5:1637; Amishiro et al . , J. Med. Chem. 1999, 42:669; Nagamura et al . , Chem. Pharm. Bull. 1996, 44:1723; Nagamura et al . , Chem. Pharm. Bull. 1995, 43; (c) CBI: Boger et al . Synthesis 1999, 1505] .

One feature limiting the attractiveness of this class of cytotoxic agents is their remarkable potencies (IC ₅₀ 5-20 pM) , creating special requirements for their preparation and handling. In many instances, this has been addressed by the introduction of chemically stable phenol protecting groups that are readily cleaved at the final stage of their preparation or upon in vivo administration. Such protected phenol precursors are intrinsically much less potent, yet readily release an active precursor to the drug upon deprotection. Extensions of this protection and release strategy have been pursued in which the free phenol release in vivo is coupled to features that might facilitate tumor selective delivery or cleavage [Wolkenberg et al . , Chem. Rev. 2002, 102:2477; Reviews on reductive activation: Papadopoulou et al . , Drugs Future 2004, 29, 807; Jaffar et al . , J. Exp. Opin. Ther. Patents 1999, 9:1371; Patterson et al . , Biomed. Health Res. 1998, 25:72}.

Such inactive prodrugs serve the dual role of providing safer handling intermediates or final products as well as potentially enhancing the

therapeutic index of the drug. As attractive and amenable as this approach is for this class of drugs, a surprisingly small series of such studies has been disclosed [Chari et al . , Cancer Res. 1995, 55:4079; Lillo et al., Chem. Biol. 2004, 21:897; Tietze et al., Eur. J. Org. Chem. 2002, 10:1634; Tietze et al . , Angew. Chem. Int. Ed. 2002, 41:759; Tietze et al . , ChemBioChem 2001, 2:758; Tietze et al . , Angew. Chem. Int. Ed. 2006, 45:6574; Wang et al . , Bioorg. Med. Chem. 2003, 11:1569; Jeffrey et al . , J. Med. Chem. 2005, 48: 1344; Kline et al . , MoI. Pharmaceut. 2004, 1:9; Hay et al . , J. Med. Chem. 2003, 46:5533; Tercel et al., J. Med. Chem. 2003, 46:2132; Gieseg et al . , Anti-Cancer Drug Design 1999, 24:77; Hay et al . , Bioorg. Med. Chem. Lett. 1999, 9:2237; Atwell et al . , J. Med. Chem. 1999, 42:3400; Atwell et al . , J. Org. Chem. 1998, 63:9414; Atwell et al . , Bioorg. Med. Chem. Lett. 1997, 7:1493; Townes et al . , Med. Chem. Res. 2002, 11:248; Boger et al . , J. Org. Chem. 1999, 69:8350] .

Alternative and prior efforts at incorporating a reductive activation into the CC- 1065 and duocarmycin class includes the Denny disclosures of nitro precursors to aryl amine variants of the phenol precursors (Hay et al . , J. Med. Chem. 2003, 46:5533; Tercel et al . , J. Med. Chem. 2003, 46:2132; Gieseg et al . , Anti-Cancer Drug Design 1999, 14; 77; Hay et al . , Bioorg. Med. Chem. Lett. 1999, 9; 2237; Atwell et al . , J. Med. Chem. 1999, 42;3400; Atwell et al., J. Org. Chem. 1998, 63:9414; Atwell et al . , Bioorg. Med. Chem. Lett. 1997, 7:1493] , Lee's use of an ester subject to cleavage upon a tethered quinone reduction (Townes et al . , Med. Chem. Res. 2002, 11:248), and some of the present inventors' report of

mitomycin- like quinone precursors to a reductively activated o-spirocyclization (versus p-spiro- cyclization) analogous to those observed with the duocarmycins or its analogues [Boger et al . , J. Org. Chem. 1999, 69:8350] .

Although the approaches have provided some increase in selectivity that result from the reductive activation, none approach that observed with mitomycin and none effectively or clearly utilize an intrinsic enzyme activity that differentiated normal versus tumor cells. Notably, it may be the ease of the mitomycin hydroquinone reoxidation to the quinone in normal cells that protects it from the effects of the drug, which occurs less readily in hypoxic tumors.

The approach detailed herein was not designed for enzymatic reductive activation, but rather for activation by cleavage of a weak N-O bond by reducing nucleophiles as shown schematically in Scheme 1, below. The expectation being that hypoxic

Scheme 1

tumor cells, with their higher reducing capacity, can carry an intrinsic higher concentration of "reducing" nucleophiles (i.e., thiols) capable of activating such derivatives making them more sensitive to the prodrug treatment [Wolkenberg et al . , Chem. Rev. 2002, 202:2477. Reviews on reductive activation: Papadopoulou et al . , Drugs Future 2004, 29:807; Jaffar et al . , J. Exp. Opin. Ther. Patents 1999, 9:1371; Patterson et al . , Biomed. Health Res. 1998, 25:72] . Moreover, as detailed below, the design lends itself to a rational tuning of the ease of reduction of the derivative, permitting empirical experience with the series to guide future design. This prodrug design also lends itself to compounds for which a slow release of active material that does not need activation is desired. These compounds do not require enzymatic release, and the release mechanism is general for other phenolic (hydroxy-

functional) drugs that can benefit from such a designed activation.

BRIEF SUMMARY OF THE INVENTION The present invention contemplates a scissile chimer molecule comprised of a targeting polypeptide portion covalently attached by a linker group to a hydroxy- functional prodrug pharmaceutical portion such as a cytotoxic or antitumor compound. In a contemplated scissile chimer, the targeting polypeptide binds to a predetermined target molecule or a portion thereof such as an epitope, and the prodrug upon cleavage from the targeting polypeptide becomes an active drug that kills the targeted cell.

A contemplated chimer molecule precursor corresponds in structure to Formula A, below, wherein

the hydroxy- functional prodrug pharmaceutical compound is represented by the circle-D portion and the hydroxyl functionality is present as part of the hydroxylamine group R ¹ R ² N-O- . In Formula A, R ¹ and R ² are the same or different and are selected from the group consisting of hydrido, C ₁ -C ₁₀ -alkyl , C7-Cio-aralkyl, C ₁ -C ₁₈ -carbonyl, ar-C ₁ -C ₁₀ - alkoxycarbonyl , C ₁ -C ₁₀ -alkoxycarbonyl, ar-C ₁ -C ₁₀ - alkylaminocarbonyl , C ₁ -C ₁₀ -alkylamino-carbony1 , C ₁ -C ₁₈ -sulfonyl, ar-C ₁ -C ₁₀ -alkoxysulfonyl,

C ₁ -C ₁₀ -alkoxysulfonyl, ar-C ₁ -C ₁₀ -alkylaminosulfonyl, and C ₁ -C ₁₀ -alkylaminosulfonyl, or R ¹ and R ² together with the depicted nitrogen atom (-NR ¹ -R ² ) form a cyclic structure that includes 4 to 6 carbon atoms in a ring .

At least one of R ¹ and R ² contains a carbonyl or sulfonyl group bonded directly to the depicted nitrogen atom. One of R ¹ and R ² also functions as the linking group for covalent attachment of the hydroxy- functional prodrug portion to a polypeptide, such as the targeting polypeptide portion of a chimer molecule. As such, that R ¹ or R ² group includes a reactive functionality (moiety) that forms a covalent bond with one or more of a mercaptan (thiol) , an amino group, a hydroxyl group, an aldehyde or a ketone group or a carboxyl group, hydroxy- functional prodrug pharmaceutical portion chimer molecule

A particularly preferred type of cytotoxic drug is a scission-activated duocarmycin-type prodrug, although other drugs such as daunomycin, doxorubicin, uncialamycin, taxol, maytansine, calichemycin, vincristine and vinblastine, and their related compounds are also contemplated. Where the prodrug is a duocarmycin-type prodrug, the released compound in activated, drug form, alkylates duplex DNA.

A contemplated scission-activated duocarmycin-type prodrug (chimer molecule precursor) can have a variety of chemical structures, but all compounds share the alkylation structural motif shown in Formula I,

In Formula I, X is a leaving group such as a halide (chloride, bromide or iodide) , a pseudohalide such as cyanide, thiocyanate, cyanate, fulminate and azide, or a sulfonic ester such as tosylate, besylate or mesylate. W is a hydrocarbyl ring or a heterocyclic ring-containing moiety or a fused ring system (radical) having five to eight atoms in the ring bonded directly to the depicted benzene ring, and containing one, two or three rings that have carbon atoms and also one to three hetero atoms heteroatoms in the ring(s) that are selected from nitrogen, sulfur and oxygen. In some preferred embodiments, the ring W is aromatic. Z is a DNA binding domain. FA and R ¹ are as discussed above.

A contemplated scissile chimer molecule can be depicted as is shown in Formula B, below, wherein

Circle-D and R ¹ are as before described, and TP-R ¹ is the targeting polypeptide covalently linked to the

former R ¹ group that upon reaction is no longer R ¹ and becomes R ³ .

A contemplated chimeric form of a scission- activated duocarmycin-type prodrug corresponds in structure to Formula IA, below, wherein W, X, Z and

R ² are as defined for Formula I, and TP-R ³ is as defined in Formula B.

A pharmaceutical composition is also contemplated. That composition contains an effective amount of the above-described scissile chimer molecules of Formula B such as a compound of Formula IA dissolved or dispersed in a pharmaceutically acceptable carrier or diluent.

A method of treating cancer cells is also contemplated. In accordance with this method, an above composition is contacted with the cancer cells to be treated. That contact can be carried out in vitro or in vivo in a mammal having a cancerous condition, and can be repeated as necessary to obtain the desired treatment effect.

The present invention has several benefits and advantages. One benefit is that a contemplated

scissile chimer molecule can be relatively benign in its form and particularly potent on activation.

An advantage of the invention is that the scissile chimer molecule can be targeted to a particular preselected cancer antigen or epitope to concentrate the cytotoxic duplex DNA alkylating portion at or near the site of the disease.

Another benefit of the invention is that the cytotoxic duplex DNA alkylating portion and associated targeting polypeptide linking group is relatively readily prepared.

Another advantage of the invention is that many contemplated targeting polypeptide portions have been prepared, are available and can be adapted for use herein by known techniques.

A further benefit of the invention is that the release mechanism is general and is readily adapted for use with a chimer molecule precursor that is a hydroxy- functional prodrug pharmaceutical compound that does not require activation to achieve its potency.

Still further benefits and advantages of the invention will be apparent to the skilled worker from the discussion that follows.

DETAILED DESCRIPTION OF THE INVENTION The present invention contemplates a scissile chimer molecule that is comprised of a targeting polypeptide portion covalently attached by a linker group to a chimer molecule precursor that is a hydroxy-functional prodrug pharmaceutical portion such as a cytotoxic or antitumor compound. The targeting polypeptide portion of a contemplated scissile chimer binds to a predetermined target

molecule or a portion thereof such as an epitope, and the prodrug upon cleavage from the targeting polypeptide, typically after endocytosis of the chimer, becomes an active drug that kills the targeted cell.

A contemplated chimer molecule precursor (hydroxy- functional prodrug pharmaceutical) portion of a scissile chimer molecule is an iV-acyl 0-amino phenol derivative that corresponds in structure to

Formula A, below, wherein the hydroxy- functional prodrug pharmaceutical compound is represented by the circle-D and the hydroxyl functionality is present as part of the hydroxylamine group R ¹ R ² N-O- . In Formula A, R ¹ and R ² are the same or different and are selected from the group consisting of hydrido, C ₁ -C ₁₀ -alkyl' C ₇ -C ₁₀ -aralkyl, C ₁ -C ₁₈ -carbonyl, ar-C ₁ -C ₁₀ -alkoxycarbonyl, C ₁ -C ₁₀ -alkoxycarbonyl, ar-C ₁ -C ₁₀ -alkylaminocarbonyl, C ₁ -C ₁₀ -alkylamino- carbonyl, C ₁ -C ₁₈ -sulfonyl, ar-C ₁ -C ₁₀ -alkoxysulfonyl, C ₁ -C ₁₀ -alkoxysulfonyl, ar-C ₁ -C ₁₀ -alkylaminosulfonyl, and C ₁ -C ₁₀ -alkylaminosulfonyl. Alternatively, R ¹ and

R ² together with the depicted nitrogen atom (-NR ¹ R ² ) form a cyclic structure that includes 4 to 6 carbon atoms in a ring. At least one, and preferably both, of RI and R ² contains a carbonyl or sulfonyl group bonded directly to the depicted nitrogen atom.

One of R ¹ and R ² also functions as the linking group for covalent attachment of the duplex DNA alkylating portion to the targeting polypeptide portion of the chiraer molecule. As such, that R ¹ or

R ² group includes a reactive functionality that can form a covalent bond with a mercaptan (thiol) , an amino group, a hydroxyl group, an aldehyde or a ketone group or a carboxyl group.

Illustrative active drug forms of prodrugs include daunomycin, doxorubicin, duocarmycins, taxol, maytansine, uncialamycin, calicheamycin, vincristine and vinblastine, and their related compounds. The chemical formulas of several of these contemplated active drugs are shown below.

R = CH ₃ = maytansine

R = CH (CH ₃ ) ₂ = maytanbutine

R = CH ₂ CH ₃ = maytanprine

R = CH ₂ CH (CH ₃ ) ₂ = maytanvaline

uncialamycin 26 - epi -uncialamycin daunomycin doxorubicin

(+)-42 dynemycin A US Pat. 5,281,710

U.S. Patent No. 5,281,710 discloses several compounds related to dynemicin A such as Compound (+) -42 that can be used herein.

As used herein, the term "related compounds" means a compound that shares an active functionality with the named compound. For example, dynemicin A and Compound (+) -42 of US Patent No. 5,281,710 are related in that they share the active functionality of a phenyl ring fused to two fused six-membered rings on of which includes a ring nitrogen atom and an ene-diyne group bridging the two fused six- membered rings. The uncialamycins depicted above are related as being R and S-isomers, and differ from the dynemicin A family of compounds in lacking the second fused six-membered ring. Similarly, maytansine, maytanbutine, maytanprine and maytanvaline all share the same active ring structure and differ only in the identity of an amidifying carboxylic acid.

A particularly preferred chimer molecule precursor hydroxy- functional prodrug pharmaceutical portion is a scission-activated duocarmycin-type prodrug. Again, the targeting polypeptide portion of a contemplated chimer binds to a predetermined target molecule or a portion thereof such as an epitope, and

the scission-activated duocarmycin-type prodrug is on cleavage from the targeting polypeptide portion of the chimer becomes an activated drug that alkylates duplex DNA.

Such phenolic precursor molecules are referred to herein as a scission-activated duocarmycin-type prodrug because cleavage of the bond between a protecting group and the phenolic oxygen leads to the formation of a corresponding duocarmycin-type compound as is shown above, i.e., (+)-CC-1065, (+) -duocarmycin A, (+) -duocarmycin SA, and (+) -yatakemycin. These prodrugs often can exist in enantiomeric forms. Although the use of each enantiomeric form is broadly contemplated, it is preferred that only one enantiomer be utilized in a chimeric molecule, as compared to a racemate being used.

A contemplated scission-activated duocarmycin-type prodrug can have a variety of Chemical structures, but all compounds share the alkylation structural motif shown in Formula I,

In Formula I, X is a leaving group such as a halide (chloride, bromide or iodide) , a pseudohalide such as cyanide, thiocyanate, cyanate, fulminate and azide, or a C ₁ -Cs sulfonic ester such as tosylate, besylate, brosylate or mesylate.

Pseudohalides are monovalent and have properties similar to those of halides [Schriver et al . , Inorganic Chemistry, W. H. Freeman & Co . , New York (1990) 406-407] . Pseudohalides include the cyanide (CN ^"1 ) , thiocyanate (SCN ^"1 ) , cyanate (OCN ^"1 ) , fulminate (CNO ^"1 ) and azide (N3 ^"1 ) anions.

W is a hydrocarbyl ring or a heterocyclic ring-containing moiety or a fused ring system (radical) having five to eight atoms in the ring bonded directly to the depicted benzene ring, and containing one, two or three rings that have carbon atoms and also one to three heteroatoms in the ring(s) that are selected from nitrogen, sulfur and oxygen. Illustrative hydrocarbyl aromatic ring groups include benzene and naphtho groups, whereas aliphatic rings include cyclopentyl, cyclohexyl, cycloheptyl and cyclooctyl that can contain one or two ethylenic unsaturations in addition to that depicted in Formula I. Illustrative heterocyclic or heteroaryl groups containing up to three heteroatoms that can be nitrogen, oxygen or sulfur include pyrrolidinyl, piperidinyl, piperazinyl, morpholinyl, thiamorpholinyl, pyrrolyl, imidazolyl (e.g., imidazol-4-yl, l-benzyloxycarbonylimidazol-4-yl, and the like), pyrazolyl, pyridyl, pyrazinyl, pyrimidinyl, furyl, tetrahydrofuryl , thienyl, triazolyl, oxazolyl, oxadiazoyl, thiazolyl, thiadiazoyl, indolyl (e.g., 2-indolyl, and the like), quinolinyl, (e.g., 2-quinolinyl, 3-quinolinyl, 1- oxido-2-quinolinyl, and the like), isoquinolinyl (e.g., 1- isoquinolinyl, 3 -isoquinolinyl, and the like), tetrahydroquinolinyl (e.g., 1,2,3,4- tetrahydro-2-quinolyl, and the like), 1, 2, 3 , 4-tetrahydroisoquinolinyl (e.g., 1,2,3,4-

tetrahydro-1-oxo-isoquinolinyl, and the like) , quinoxalinyl, β-carbolinyl, 2-benzofurancarbonyl, benzothiophenyl, 1-, 2-, 4- or 5-benzimidazolyl, and the like radicals. W is preferably a benzo or a pyrrolo moiety so that the resulting duplex DNA alkylating portion of the chimer corresponds in structure to one or the other of Formulas HA or UB, below, wherein X, Z, R ¹ and R ² are defined here.

More particularly, Z is a DNA binding domain whose structure can be extremely varied. Illustrative DNA binding domains are exemplified by the binding domains of (+)-CC-1065, (+) -duocarmycin A, (+) -duocarmycin SA, and (+) -yatakemycin that were shown before. Further illustrative DNA binding domains are those of Compounds 4 and 8 disclosed

herein, as well as the domains of bleomycin that are shown linked to the duplex DNA alkylating domain of duocarmycin-type prodrugs in US Patent No. 6,060,608 to Boger.

As noted above, R ¹ and R ² are the same, or preferably different. At least one, and preferably both of R ¹ and R ² , are C ₁ -C ₁₀ carbonyl or sulfonyl

groups bonded directly to the depicted nitrogen atom, thereby forming a substituted hydroxamate [-C(O)NO-] or a hydroxysulfonamide [-S(O) 2NO-] . An R ¹ or R ² group can also be a hydrido, C ₁ -C ₁₀ alkyl, C ₇ -C ₁₀ - aralkyl, ar-C ₁ -C ₁₀ -alkoxycarbonyl, C ₁ -C ₁₀ " alkoxycarbonyl , ar-C ₁ -C ₁₀ -alkylaminocarbonyl, or C ₁ - C ₁₀ -alkylaminocarbonyl, or R ¹ and R ² together with the depicted nitrogen atom (-NR ¹ R ² ) form a cyclic structure that includes 4 to 6 carbon atoms in a ring. In one preferred embodiment, R ¹ and R ² together with the depicted nitrogen atom form a cyclic structure shown below in Formula III

One of R ¹ and R ² also functions as the linking group for covalent attachment of the duplex DNA alkylating portion to the target polypeptide portion of the chimer molecule. As such, that so functioning R ¹ or R ² group includes a reactive functionality spaced away from the depicted nitrogen atom that can form a covalent bond with a mercaptan (thiol) , an amino group, a hydroxyl group, an aldehyde or a ketone group or a carboxyl group. Illustrative covalent bond- forming groups include a thiol, a Michael addition accepting α,β-unsaturation, a carboxylic acid, a hydrazide or semicarbazide, or an amine. The -NR ¹ R ² group of Formula III reacts with a thiol in a Michael addition that can link the two portions of the chimer molecule .

In any of the Formulas herein, the term "C ₁ -C ₁₀ alkyl" denotes a straight or branched chain radical such as a methyl, ethyl, n-propyl, isopropyl, n-butyl, sec-butyl, tert-butyl, amyl, tert-amyl, hexyl, heptyl, decyl group and the like. The term "lower alkyl" denotes a C ₁ -C ₄ alkyl group. A preferred "C ₁ -C ₁₀ alkyl" group is a methyl group.

The term "C ₇ -C ₁₀ -aralkyl" or "C ₇ -C ₁₀ phenylalkyl" denotes a C ₁ -C ₄ alkyl group substituted at any position by a phenyl ring. Examples of such a group include benzyl, 2-phenylethyl, 3 -phenyl (n-prop- 1-yl) , 4 -phenyl (hex- 1-yl) , 3 -phenyl (n-am-2-yl) , 3 -phenyl (sec-butyl ), and the like. A preferred C ₇ -C ₁₀ phenylalkyl group is the benzyl group.

The term "C ₁ -C ₁₈ -carbonyl" denotes a group of the formula RC(O)- where R = C ₁ -C ₁₈ -alkyl such as methyl, ethyl, propyl, butyl, sec-butyl, hexyl, heptyl, octyl, nonyl, decyl, lauryl, myristyl, palmityl and stearyl so that the substituent group is an acyl group such as formyl, acetyl, propionyl, hexanoyl, lauroyl, stearoyl and the like.

The term "ar-C ₁ -C ₁ o-alkoxycarbonyl" denotes a group of the formula R-O-C(O)- where R = C ₁ -C ₁₀ - alkyl in which the alkyl group is substituted at any position by a phenyl ring, as above. This nomenclature is used to highlight the possibly longer chainl length of the alkyl group portion of the substituent as compared to that used above.

The term "C ₁ -C ₁₀ -alkoxycarbonyl" denotes a group of the formula R-O-C(O)- where R = C ₁ -C ₁₀ -alkyl in which the alkyl group is as discussed abvoe .

The term "C ₁ -C ₁₀ -alkylaminocarbonyl" denotes a group of the formula R-NH-C(O)- where R = C ₁ -C ₁₀ -alkyl in which the alkyl group is as discussed abvoe .

The term "ar-C ₁ -C ₁₀ -alkylaminocarbonyl" denotes a group of the formula R-NH-C(O)- where R = C ₁ -C ₁₀ -alkyl] in which the alkyl group is substituted at any position by a phenyl ring, as above.

The term "C ₁ -C ₁₈ -sulfonyl" denotes a group of the formula R-S (0)2- where R = C ₁ -C ₁₈ -alkyl in which the alkyl is as discussed before and further includes long chained alkyl group such as heptyl, octyl, nonyl, decyl, lauryl, myristyl, palmityl and stearyl .

The term "C ₁ -C ₁₀ -alkoxysulfonyl" denotes a group of the formula R-O-S (0)2- where R = C ₁ -C ₁₀ - alky1 in which the alkyl is as discussed before.

The term "ar-C ₁ -C ₁₀ -alkoxysulfonyl" denotes a group of the formula R-O-S (0)2- where R = C ₁ -C ₁₀ - alkyl in which the alkyl is as discussed before and that alkyl group can be substituted at any position with a phenyl group.

The term "C ₁ -C ₁₀ -alkylaminosulfonyl" denotes a group of the formula R-NH-S (0)2- where R = Ci-Cio -alkyl in which the alkyl is as discussed before .

The term "ar-C ₁ -C ₁₀ -alkylaminosulfonyl" denotes a group of the formula R-O-S (0)2- where R = C ₁ -C ₁₀ -alkyl in which the alkyl is as discussed before and that alkyl group can be substituted at any position with a phenyl group.

TARGETING POLYPEPTIDES

A contemplated targeting polypeptide is itself soluble in aqueous media that are used for i.v. or i.p. administrations. It is therefore preferred that the targeting polypeptide be a secreted molecule or that the extracellular portion of a membrane-bound protein be utilized as a targeting polypeptide. Antibody molecules are exemplary of contemplated secreted targeting polypeptides, whereas the extracellular portions of membrane-bound molecules such as the E-, P- and L- selectin molecules exemplify targeting polypeptides whose intact protein is not soluble in appropriate aqueous media. The extracellular portion of E-selectin is a preferred targeting polypeptide, and the portion that binds to the silayl Lewis ^x and silayl Lewis ^a antigens is more preferred.

Idiotype-containing polyamide portions (paratopes or antibody combining sites) of antibodies are particularly preferred targeting polypeptides. Those polypeptides are portions of antibody molecules that include the idiotype, and bind to a target epitope (ligand) . Such portions include the Fab, Fab' and F(ab')2 fragments prepared from antibodies by well-known enzymatic cleavage techniques. See for example, U.S. Patent No. 4,342,566 to Theofilopoulos and Dixon, generally, and specifically, Pollack et al., Science 1986 234:1570-1573, who reported accelerated hydrolytic rates for Fab fragments were the same as those of the native Ig.

Antibody Fc portions bind to receptors on several types of lymphocytes such as B cells, various T cells, and neutrophils. Thus, the Fc portions of

antibodies can be used to target the chimer molecules to a cancerous lymphocyte .

Inasmuch as the antibodies from which idiotype-containing polyamides are obtained are described as raised against or induced by immunogens, idiotype-containing polyamide receptors can also be discussed as being "raised" or "induced" with the understanding that a cleavage step is usually required to obtain an idiotype-containing polyamide from an antibody. Intact antibodies are preferred, however, and are utilized herein as illustrative of the targeting polypeptide molecules of this invention.

The targeting polypeptides useful in the present invention are preferably monoclonal antibodies. A "monoclonal antibody" is an antibody produced by clones of a single cell called a hybridoma that secretes but one kind of antibody molecule that binds to a target ligand. The hybridoma cell is fused from an antibody-producing cell and a myeloma cell or other self-perpetuating cell line.

Techniques for preparing the monoclonal antibodies of the present invention are well known. Such receptors were first described by Kohler and Milstein, Nature, 1975 256:495-497. Monoclonal antibodies are typically obtained from hybridoma tissue cultures or from ascites fluid obtained from mammals into which the hybridoma tissue was introduced.

Monoclonal antibodies (mAbs) are preferred herein because of their unique specificity in binding to a single epitope of a target ligand, as well as their relatively higher specific activities as

compared to polyclonal antibodies. Polyclonal antibody preparations can also be used herein.

The mAb utilized immunoreacts substantially only with a single target epitope of a tumor cell; i.e., is tumor cell specific, and thereby provides further specificity to the drug molecules . Such a Mab- linked fused ring prodrug is one type of chimeric molecule of the invention.

The mAb portion of the above chimeric construct can constitute an intact antibody molecule of IgG or IgM isotype, in which case, a plurality of compounds can be present per antibody molecule . The binding site portions of an antibody can also be utilized, in which case, at least one compound is linked to the proteinaceous antibody binding site portion.

An antibody binding site portion is that part of an antibody molecule that immunoreacts with an epitope, and is also sometimes referred to as a paratope. Binding site-containing portions of antibodies can be referred to as paratope-containing molecules .

Exemplary anti- tumor Mabs are noted in the table below, listed by the name utilized in a publication, along with its deposit accession number at the American Type Culture Collection (ATCC) , 12301 Parklawn Drive, Rockville, Md. 20852 U.S.A., and the tumor antigen with which the Mab paratope is reported to react. A citation to a discussion of each Mab and its immunoreactivity is provided by the footnote under the antigen listing.

PRODRUG LINKERS CHEMISTRY

General Synthesis

The discussion immediately below provides illustrations of the general chemistry involved here

and specific reaction types that are useful herein. The discussion that follows in the next section on "Linker Groups" relates more specifically to the present invention.

A range of methods for direct conversion of a precursor phenol to the corresponding 0-amino phenol were examined (O-amidation) and several routes to the final compounds were explored. It was anticipated that this might best be conducted on a seco-N-Boc-CBI derivative lacking the capabilities of spirocyclization (e.g., Compound 11) . However, the lability of the resulting N-acyl 0-amino phenol derivatives to subsequent chemical transformations proved significant and this approach proved less viable than a surprisingly effective direct O-amidation reaction of seco-CBI-TMI or seco-CBI- indole ₂ that is shown in Schemes 2A and 2B, below. Thus, low temperature phenol deprotonation of Compound 2 {3 equiv of lithium hexamethyldisilazane

Scheme 2A

[LiHMDS; Li (CH ₃ ) ₃ SiNHSi (CH ₃ ) ₃ ] , zero ⁰ C, ether- dioxane} followed by treatment with the amidating reagents TsONHBoc [Greek et al . , Bull. Soc. Chim. Fr. 1994, 131:429] or TsONPhth (Neumann et al . , J. Biol. Chem. 1994, 269:21561] provided Compounds 4 and 7 directly in good conversions. Similar results are had using the corresponding maleimido derivative, TsONHMaI. Competitive spirocyclization of Compound 2 to CBI-TMI itself was observed if the deprotonation was carried out at higher reaction temperatures or in more polar solvents. Spirocyclization of Compound 2 diminished as the solvent polarity was reduced (glyme > THF > dioxane-ether > ether, insoluble) and was less prominent with LiHMDS versus NaHMDS.

In most instances, recovered starting phenol was present in the crude reaction product and was chromatographically close enough to the iV-acyl 0-amino phenols that special precautions were taken to ensure its removal. Those precautions entailed exposure of the product mixture to conditions that promote deliberate spirocyclization of the seco phenol derivatives (saturated aqueous NaHCO ₃ -THF (1:1), 23 ⁰ C, 2 hours) [Boger et al . , J. Org. Chem. 1992, 57:2873; Boger et al . , J. Org. Chem. 1995, 50:1271; Drost et al . , J. Org. Chem. 1991, 55:2240; Aristoff et al . , J. Org. Chem. 1992, 57:6234; Mohamadi et al . , J. Med. Chem. 1994, 37:232; Ling et al . , Heterocyclic Commun. 1997, 3:405; Boger et al . , Synlett 1997, 515; Boger et al . , Tetrahedron Lett. 1998, 39:2227; Kastrinsky et al . , J. Org. Chem. 2004, 59:2284] and subsequent chromatographic separation of the much more polar CBI-TMI or CBI-indole ₂ .

AJ-Acetylation of Compound 4 (Ac ₂ O, cat. DMAP, CH ₂ Cl ₂ , 23 ⁰ C, 12 hours, 81%) provided Compound 6 and subsequent Boc deprotection (TFA-CH ₂ Cl ₂ (1:1), 23 ⁰ C, 3 hours, 88%) afforded Compound 5. In an analogous manner, seco-CBI-indole ₂ (Compound 3) was directly converted to Compound 8 (45%) upon LiHMDS deprotonation (3 equiv of LiHMDS, ether-dioxane , zero ⁰ C, 30 minutes) and subsequent O-amidation with TsONHBoc [Greek et al . Bull. Soc. Chim. Fr. 1994, 232:429] .

Scheme 2B

For comparison purposes, two analogues of seco-CBI-TMI were prepared that are incapable of spirocyclization to CBI-TMI itself. The first incorporates the C4 phenol protected as its methyl ether (Compound 10) and second contains no C4 substituent (Compound 9) . The former was prepared from Compound 11 (Kastrinsky et al . , J. Org. Chem. 2004, 59:2284) by phenol O-methylation, primary alcohol OTBS deprotection and subsequent conversion to the primary chloride Compound 14, followed by N-Boc deprotection and coupling with 5,6,7- trimethoxyindole-2-carboxylic acid (Compound 15) to provide Compound 10, as is shown in Scheme 3, below.

Scheme 3

Throughout this sequence and as a result of the multiple purifications, the chances of residual, contaminant phenol (Compound 2) being present in the final product Compound 10 are remote. Nonetheless, because even trace quantities of Compound 2 can be misleadingly detected in the subsequent biological evaluations (e.g., 0.01%), the analogue Compound 9 was also prepared for comparison and by an approach that precludes the presence of such a contaminate phenol .

Thus, following a route analogous to that used for CBI itself (Boger et al . , J. Org. Chem. 1992, 57:2873; Boger et al . , J. Org. Chem. 1995, 60:1271; Drost et al . , J. Org. Chem. 1991, 56:2240; Aristoff et al . , J. Org. Chem. 1992, 57:6234; Mohamadi et al . , J. Med. Chem. 1994, 37:232; Ling et al., Heterocyclic Commun . 1997, 3:405; Boger et al . , Synlett 1997, 515; Boger et al . , Tetrahedron Lett. 1998, 39:2227; Kastrinsky et al . , J. Org. Chem. 2004,

59:2284), Compound 20 was prepared from Compound 16 and converted to Compound 21, enlisting a key 5-exo- trig aryl radical-alkene cyclization [Boger et al . , Tetrahedron Lett. 1998, 39:2227] , as is shown in Scheme 4, below.

Scheme 4

c

The product Compound 21, like Compound 14 (α = 1.19), was chromatographically resolved on a semipreparative ChiralCel OD column (α = 1.42) providing each enantiomer, and Compound 21 was coupled with 5, 6, 7-trimethoxyindole-2-carboxylic acid

(Compound 15) upon W-Boc deprotection to provide Compound 9.

Linker Groups

Methods for linking prodrugs to targeting polypeptides such as antibodies are well known in the art. Antibodies and other proteins contain reactive side groups such as carboxylic acids from aspartic and glutamic acids, a primary amine from lysine, a mercaptan from cysteine or cystine, and hydroxyl groups from serine, threonine and tyrosine. Well known chemistries have been developed for covalently bonding (linking) pendant molecules from each type of reactive side group. Whole antibodies are used illustratively herein as targeting polypeptides.

Illustratively, mercaptan groups (-SH) can be prepared in an antibody by reduction of the disulfides present. For example, monoclonal antibody (mAb) 6D2 was prepared from mice immunized with melanin produced by the fungus, Cryptococcus neoformans [Rosas et al . , Infect. Immun. (2000) 68:2845-2853] . Mercaptan (-SH) groups are prepared on mAb 6D2 by reduction with dithiothreitol (DTT) using 0.5 mg of 6D2 as reported in Dadachova et al . , Proc. Natl. Acad. Sci . USA (2004) 101:14865-14870. The antibody samples are purified from unreacted reducing agents on Centricon-50 microconcentrators by washing with 3 x 1.5 mL of ammonium acetate buffer, pH = 6.5-7.0. The determination of -SH groups on the antibody is carried out using Ellman's reagent with spectrophotometric detection according to Ellman's reagent manufacturer instructions (Pierce, USA) and techniques described in Dadachova et al . , Nucl. Med. Biol. (1997) 24(6) :605-608.

The number of -SH groups per 6D2 molecule is determined to be about 56 for a DTT reduction. Processing the mAb with DTT results in some fragmentation of the antibody.

A reduced antibody so prepared can thereafter be reacted with a mercaptan-active linking agent such as a maleimide that undergoes a Michael addition reaction with mercaptans under mild conditions. Illustrative preparations of contemplated prodrugs such as CBI-TMI and CBI-indole ₂ are illustrated hereinafter in Schemes 5 and 6, respectively, for linkage to targeting polypeptide mercaptan groups, whereas the reactions of Schemes 7 and 8 are designed to prepare a prodrug for linkage to an oxidized glycosyl group of an antibody or other targeting polypeptide (Scheme 7) , for linking to an amine or alcohol group (Scheme 8) , reactions of Scheme 9 can be used to covalently attach a prodrug to a targeting polypeptide amine group, whereas the reactions of Scheme 10 provide a means for linking to a targeting polypeptide carboxylic acid group

Scheme 5 shows a reaction sequence in which CBI-TMI (Compound 2) is initially reacted as shown in Scheme 2A. Thus, Compound 2 is reacted to form Compound 4 as discussed above. That reaction is followed by reaction with N- [ε-maleimidocaproyloxy] - succinimide ester (EMCS) or a similar amidating reagent such as N- [γ-maleimidobutyryloxy] succinimide ester (GMBS) or N- [β-maleimidopropyloxy] succinimide ester (BMPS) to form the N-Boc hydroxamate Compound 16 (or similar compound using GMBS or BMPS, all three available from Pierce Chemical) , from which the Boc group can be readily removed with TFA to form Compound 17, that itself can be acylated as with

acetic anhydride or another acylating agent to form Compound 18 (or similar compound using GMBS or BMPS) . Acylation with acetic anhydride or similar group further activates the N-O bond toward cleavage and displacement of the chloride group from across the molecule.

The reaction sequence shown in Scheme 6 is similar to that shown in Scheme 2B for Compound 8 with the addition of a further reaction that further activates the N-O bond toward cleavage. Here, a urea-hydroxamate is formed as Compound 18 using

N- [p-maleimidophenyl] isocyanate (PMPI; Pierce Chemical) . The Boc-protecting group can be left in place or removed before or preferably after the compound is reacted with the mercapto groups of a reduced antibody to form the Michael product using mild acid treatment.

As noted by Abraham et al . , J Immunol Methods. 1991 144 (1) :77-86 , about 10 to about 25 aldehyde groups can be generated on an IgG antibody by mild oxidation with periodate, whereas more than 200 aldehydes could be prepared on an IgM antibody

under the same conditions. Some binding avidity of the antibodies studied was said to be lost when the periodate concentration exceeded about 50 mM and was more pronounced if the oxidation were carried out at pH 5.6 and 25° C as compared to pH 4.6 and zero degrees C.

Scheme 7, below, illustrates an exemplary- synthetic sequence for the preparation of a prodrug adapted to be reacted with (bound to) an oxidized sugar moiety of an antibody or other targeting polypeptide. Here, again in initial reactions analogous to those shown in Scheme 2A, a prodrug such as Compound 2 is reacted to form Compound 4 as discussed above. The N-protected Compound 4 is then reacted with succinimidyl 4-hydrazinonicotinate acetone hydrazone (SANH) or the similar compound that is referred to in the literature as C6-SANH, both of which are available from Pierce Chemicals and whose structures are shown in Scheme 7. That reaction forms a hydroxamate ester Compound 20 (R ² = ANH) from which the Boc group can be readily removed to form Compound 22 that can be further acylated as with acetic anhydride to form the even more reactive prodrug Compound 24 or with another acylating agent. Removal of the reacted acetone blocking group provides a hydrazine compound that can be reacted with an oxidized glycosyl group of protein such as an antibody to provide the completed targeted prodrug.

Scheme 7

Yet another preparation of a contemplated prodrug for linkage to a targeting polypeptide is shown in Scheme 8, below. There, again using Scheme 2A as a basis, Compound 4 is formed as already discussed. Reaction of Compound 4 with a cyclic anhydride such as succinic (SUC) , gluratic, adipic, pimelic anhydrides shown in Scheme 8, or an even longer dicarboxylic acid anhydride such as azeleic anhydride or docecanoic anhydride forms hydroxamate Compound 26 having a free carboxyl group that can be used to link the prodrug to an amine or hydroxy1 group of a targeting polypeptide. Again, removal of the Boc protecting group with an acid such as TFA

provides Compound 28, that can be acylated with acetic anhydride as shown to form Compound 30 or with another acylating agent as may be desired.

Scheme 8

A still further synthesis of a contemplated prodrug for bonding to a targeting polypeptide is shown in Scheme 9, below. There, again using Scheme 2A as a basis, Compound 4 is formed as already discussed. Reaction of Compound 4 with an activated N- (iodoacetamido) benzoate such as N-succinimidyl (4- iodoacetyl) aminobenzoate (SIAB; Pierce Chemical) to form a hydroxamate Compound 32 having an iodoacetamido group that can be used to link the prodrug to an amine group of a targeting polypeptide. Again, removal of the Boc protecting group with an acid such as TFA provides Compound 34, that can be

acylated with acetic anhydride as shown to form Compound 36 or with another acylating agent as may be desired.

Scheme 9

_ = IAB

A still further synthesis is shown in Scheme 10, below, for preparation of a contemplated prodrug for linking to a targeting polypeptide. Here, using Scheme 2B as an illustrative basis, Compound 3 is used to form Compound 8 as already discussed. Reaction of Compound 8 with TFA removes the t-Boc group to provide Compound 40. The latter compound is reacted with an α-N-oxysuccinimido-C2- Ci3 -CD-N-t-Boc-alkyleneamine to form Compound 42 from which the t-Boc protecting group can be removed by another treatment with TFA to form Compound 44.

Scheme 10

It is noted in any of the syntheses of Schemes 5-10 that the t-Boc can be removed prior to acylation with the linking group- containing moiety.

A contemplated scissile chimer molecule that comprises a targeting polypeptide covalently linked to the linking portion of substituent R ¹ to form an R ³ substituent a can be depicted as is shown in Formula B, below, wherein circle -D and R ² are as

before described, and TP-R ³ is the targeting polypeptide covalently linked to the former R ¹ group that upon that linking reaction is no longer R ¹ , and becomes R ³ ; i.e., the reaction product of TP and R ¹ . For example, if a mercaptan-active linking agent such as a maleimide is incorporated into the substituent

R ¹ group, Michael addition of the mercaptan across the maleimido double bond alters the R ¹ group maleimide and therefore that group is renamed as R ³ .

A contemplated chimeric form of a scission- activated duocarmycin-type prodrug corresponds in structure to Formula IA, below, wherein W, X, Z and

R ² are as defined for Formula I, and TP-R ³ is as defined in Formula B.

Pharmaceutical Compositions and Treatment Methods

A pharmaceutical composition for treating cancer cells, is also contemplated. Such a composition contains a pharmaceutically effective amount of a before-discussed chimer molecule dissolved or dispersed in a pharmaceutically acceptable diluent.

A contemplated chimer molecule can be used in a pharmaceutical composition to treat and preferably kill cancer cells in vitro or in vivo in a mammalian subject. Thus, an above composition is contacted with the cancer cells to be treated. The cells so treated are maintained in contact with the chimer molecules until cleared by the body when in vivo, or for various times as desired in an in vitro study. The treatment is generally repeated several times .

A mammal to which or whom a chimer molecule composition is administered can be a primate such as a human, an ape such as a chimpanzee or gorilla, a monkey such as a cynomolgus monkey or a macaque, a laboratory animal such as a rat, mouse or rabbit, a companion animal such as a dog, cat, horse, or a food animal such as a cow or steer, sheep, lamb, pig, goat, llama or the like in need of treatment for a cancerous condition.

A contemplated composition is administered to a mammal in need of the medication at an anticancer effective dosage level. That level is typically an amount sufficient to provide about 10 to about 100 μg/kg of body weight to the recipient's plasma or serum, using the molecular weight of the scission-activated duocarmycin-type prodrug Compound 8 itself as the basis for calculation in view of the

different molecular weights of the other prodrug compounds contemplated herein. The amount can vary depending on the recipient and cancer load. Those plasma or serum concentrations can usually be obtained by i.v. administration using a liquid dosage form that contains about 200 mg to about 1000 mg of chimer compound per day. The determination of optimum dosages for a particular situation is within the skill of the art.

Without wishing to be bound by theory it is believed that the targeting polypeptide binds to a target molecule on the surface of a cancer cell and is then brought into the by endocytosis or similar mechanism. Once within the cell, the prodrug is cleaved from the chimer by a reductive and/or nucleophilic reaction, and is thereby activated to alkylate the cell's duplex DNA, and thereby kill the cell.

A chimer molecule composition is administered repeatedly, on a schedule adapted for a recipient's cancer load and need, as is well known in the art. Typical administrations are given multiple times within a one month time period, usually followed by a rest period and then further administrations and rest periods until the recipient is free of the disease, or longer for prophylactic purposes .

For preparing pharmaceutical compositions containing a chimer compound of the invention, an inert, pharmaceutically acceptable carrier or diluent is used. The diluent is usually in liquid form.

Liquid pharmaceutical compositions include, for example, solutions suitable for parenteral administration. Sterile water solutions of the active chimer or sterile solutions of the active component in solvents comprising water, ethanol, or propylene glycol are examples of liquid compositions suitable for parenteral administration.

Sterile solutions can be prepared by dissolving the active component in the desired solvent system, and then passing the resulting solution through a membrane filter to sterilize it or, alternatively, by dissolving the sterile compound in a previously sterilized solvent under sterile conditions .

Preferably, the pharmaceutical composition is in unit dosage form. In such form, the composition is divided into unit doses containing appropriate quantities of the active urea. The unit dosage form can be a packaged preparation, the package containing discrete quantities of the preparation, for example, in vials or ampules.

ILLUSTRATIVE STUDIES

Stability and Reactivity of the N-Acyl O-Amino Phenol Derivatives

Clear from efforts directed at their preparation, the N-acyl amino phenol prodrugs displayed a useful range of stability, yet were susceptible to cleavage of the critical N-O bond. As

might be anticipated, their relative stability followed the order of Compound 4 > Compound 5 > Compound 6 > Compound 7 , with Compound 4 and Compound 5 withstanding even long term storage effectively, but with Compound 7 noticeably deteriorating over time .

Derivatives Compound 4 and Compound 6, as well as Compound 7, proved surprisingly robust to acidic conditions (TFA-CH ₂ Cl ₂ , 4 N HCl-EtOAc) , and stable to mild base treatment in nonpolar, aprotic solvents (Et ₃ N or DMAP, CH ₂ Cl ₂ ) , but exhibited a diminished stability as the solvent polarity increases: stable to NaHCO ₃ in THF or THF-H ₂ O, but cleaved in NaHCO ₃ /DMF-H ₂ O or H ₂ O and DBU/CH ₃ CN. Similarly, Compound 4 proved stable in MeOH, but Compound 2 was released slowly upon treatment with

NaHCO ₃ or Na ₂ CO ₃ in MeOH (2 hours, 23 ⁰ C) .

Most pertinent to the potential source of cleavage under physiological conditions, Compound 4 was stable to treatment with BnSH in THF (2-72 hours, 23 °C) or MeOH (2-72 hours, 23 °C) , and stable to treatment with BnSH in THF even in the presence of insoluble NaHCO ₃ (2 hours, 23 °C) . However, Compound 4 was cleaved to release Compound 2 upon treatment with BnSH in MeOH in the presence of NaHCO ₃ (2 hours, 23 ⁰ C) . Significantly, the stability of Compound 4 was assessed in pH 7.0 phosphate buffer and within the limits of detection (HPLC, UV) , no significant cleavage of the prodrug was observed over the time monitored (72 hours) . Finally, the stability of Compound 4 was monitored in human plasma (50 μg/100 μL, 10% DMSO) in which it displayed a half-life of 3 hours with release of the free drug Compound 2.

Biological Properties

Cytotoxic Activity

The 0-amino phenol derivatives bearing the N-O prodrug linkages and the various λJ-acyl substituents were assayed for cytotoxic activity alongside the parent drugs CBI-TMI (Compound 2) [Boger et al . , J. Am. Chem. Soc . 1994, 225:7996] and CBI-indole ₂ (Compound 3) [Boger et al . , Bioorg. Med. Chem. Lett. 1991, 2:115; Boger et al . , Bioorg. Med. Chem. 1995, 3:1429] as well as the two control standards Compound 9 and Compound 10 incapable of free phenol release. Three cell lines were examined including a standard L1210 cell line (mouse leukemia) as well as the mitomycin-sensitive (H460, expresses high levels of DT-Diaphorase) and resistant (H596, lacks DT-Diaphorase) non small cell lung cancer (NSCLC) cell lines.

Several important trends emerged from these studies. First, the natural enantiomer control standards Compound 9 and Compound 10, incapable of free phenol release, were inactive against all three cell lines (IC ₅₀ >100 nM) being >10, 000-fold less active than the free drug Compound 2 (seco-CBI-TMI) , as is seen from the Tables below.

Natural enantiomer series

Unnatural enantiomer series

In sharp contrast, the natural enantiomers of the 0-amino phenol prodrugs exhibited potent cytotoxic activity approaching that of the free drug itself (1-0.1 times the activity of Compound 2),

indicating its successful release under the assay conditions. Even more significantly, the relative potency of the prodrugs, when distinguishable, mirrors the expected ease of N-O bond cleavage (e.g. L1210: Compound 7 > Compound 6 > Compound 5 > Compound 4) , suggesting fundamental chemical principles may be used to "tune" the reductive free drug release.

Provocatively, the potency differences between the free drug Compound 2 and the prodrugs diminish as the hypoxic character of the cell line increases; Compound 4 is 10 -fold less potent than Compound 2 against L1210, but Compound 2 and Compound 4 are essentially equipotent against H460/H596. More significantly and unlike mitomycin C, this reductive activation is not linked to the expression levels of DT-Diaphorase because Compound 2 and Compounds 4-7 remain equipotent in the H460 or H596 cell lines, although H596 is 10-fold less sensitive than H460 to seco-CBI-TMI itself. This illustrates that DT- Diaphorase is not mediating the reductive release of the drug from the 0-amino phenol prodrugs indicating that their utility is orthogonal to that of mitomycin. Rather, their behavior is consistent with the suggestion that the activation is nonenzymatic and likely is mediated in situ by appropriate nucleophiles. Analogous trends are also observed with the CBI-TMI unnatural enantiomers albeit at concentrations that are approximately 100 to 1000- fold higher than that of the natural enantiomers.

Especially interesting and exciting was the behavior of the CBI-indole ₂ prodrug. For this CBI analogue, only the NHBoc derivative was examined

because it was the most stable of the IV-acyl 0-amino phenol prodrugs examined (tables below) . In each

Natural enantiomer series

IC ₅₀ (nM)

Compd, R

L1210 H460 H596 mitomycin C 40 20 5000

3, OH 0.03 0.2 2

8, ONHBoc 0.05 0.3 4

Unnatural enantiomer series

IC ₅₀ (nM)

Compd, R

L1210 H460 H596 mitomycin C 40 20 5000 3, OH 0.7 6 40

8, ONHBoc 2 10 60

cell line examined, the prodrug Compound 8 was essentially equipotent with CBI-indole ₂ itself, indicating effective release of the free drug under the conditions of the assay. In addition, it proved to be exceptionally potent being 100-1000 times more active than mitomycin C (IC ₅₀ = 30-200 pM vs 20-40 nM) and it remained remarkably active against the mitomycin-resistant H596 cell line (IC ₅₀ = 4 nM vs 5

μM) . Even the unnatural enantioraer of Compound 8, which was found to be 10-100 fold less active than the natural enantiomer, proved to be more active than mitomycin C. Given the efficacy of (+) -CBI-indole ₂ in animal tumor models, [Boger et al . , Bioorg. Med. Chem. Lett. 1991, 1:115-120; Boger et al . , Bioorg. Med. Chem. 1995, 3:1429-1453] it was especially interesting to compare Compound 8 with Compound 3 in vivo .

DNA Alkylation Selectivity and Efficiency

The DNA alkylation properties of Compound 4 were examined alongside the parent drug CBI-TMI (Compound 2) , and the two control standards Compound 9 and Compound 10 (incapable of spirocyclization) within w794 duplex DNA [Boger et al . , Tetrahedron 1991, 47:2661] for which results for an extensive series of duocarmycin analogues have been reported. The sites of DNA alkylation and its efficiency were directly assessed by thermally- induced singly 5' end- labeled duplex DNA strand cleavage following incubation with the agents (Fig. 1 natural enantiomers examined) .

The reductively activated agent Compound 4 was found to alkylate w794 DNA with an identical sequence selectivity as the parent agent CBI-TMI (Compound 2) , albeit with a substantially reduced efficiency (1,000-10,000 fold) . Similarly, the O-methyl ether Compound 10 as well as Compound 9 lacking a C4 substituent failed to exhibit significant observable DNA alkylation. In fact, Compound 9 showed no appreciable DNA alkylation even under forcing conditions (37 ⁰ C, 18 hours, data not shown) , whereas the potentially more reactive

O-methyl ether Compound 10 (via assisted phenonium ion formation) displayed perhaps a trace amount of DNA alkylation (<0.01% that of Compound 2) that could be attributed to either its direct, but much less facile, DNA alkylation or contaminant free phenol present in the synthetic sample of Compound 10.

With detection of DNA alkylation by the prodrug Compound 4 at the level observed (0.1-0.01% of 2) , it could not be distinguished whether this is due to direct alkylation by Compound 4 itself, trace release of Compound 2 from Compound 4 under the DNA incubation conditions {in situ N-O cleavage) , or attributable to trace contaminate Compound 2 in the synthetic samples of Compound 4. What the results do indicate is that Compound 4 is incapable of significant DNA alkylation in its own right (requires N-O bond cleavage) , and that Compound 4 is essentially stable to the DNA alkylation conditions examined, requiring deliberate N-O bond cleavage to initiate effective DNA alkylation.

These observations are consistent with the stability of Compound 4 observed in pH 7.0 phosphate buffer. Significantly, the results then suggest that the in vitro cytotoxic activity of Compound 4, and by analogy that of the related 0-amino phenol prodrugs which all approach that of the parent drug CBI-TMI (Compound 2) , is derived from in situ intracellular cleavage of the N-O bond and productive release of the active drug under the cell culture conditions .

In Vivo Antitumor Activity

The prodrug Compound 8 was examined for in vivo efficacy alongside the parent drug Compound 3 in a standard antitumor model enlisting L1210 murine

leukemia implanted i.p. into DBA/2J mice. This model has been reported to respond well to the parent drugs of related compounds [Li et al., Invest. New Drugs 1991, 9.-137) and is a system that collaborators through the years have used to assess an extensive series of (+) -CBI-indole ₂ analogues. Although not published, these latter studies provided the foundation on which examination of Compound 8 was based.

With use the dose range (10-100 μg/kg) and the dosing schedule (administered three times i.p. on days 1, 5, and 9) found suitable for related parent drugs including (+) -CBI-indole ₂ (Compound 3) [Boger et al., Bioorg. Med. Chem. Lett. 1991, 2:115-120; Boger et al., Bioorg. Med. Chem. 1995, 3:1429-1453] , the prodrug Compound 8 was examined (Table below) . The dose at which a maximal response was observed for

Compound 8 corresponded closely to that of (+) -CBI- indole ₂ (Compound 3) , although its efficacy was significantly improved. This result indicates that the prodrug Compound 8 (1) efficiently and effectively releases the free drug Compound 3 in the in vivo model (reductive activation) , and (2) that either the rate of release or the site of release enhances the efficacy of the drug.

Moreover, the efficacy of Compound 8 is extraordinary, providing 5/6 long-term survivors at 52 weeks (365 days, T/C >1550) at the optimal dosing examined (100 μg/kg) , and that continue to be monitored. Notably, little distinction between Compound 3 and Compound 8 was observed at days 30-100 except that the prodrug-treated animals appeared healthier, displaying little or no weight loss which was evident with Compound 3 at the highest dosing.

With the prolonged management of the treated animals herein that exceeded the time frame typically allotted for such an in vivo antitumor assessment, it was observed that the surviving mice at day 90 treated with the free drug Compound 3, but not the prodrug Compound 8, eventually expired due to drug administration related complications. (This result appears to arise from damage to the intraperitoneal cavity or its organs that originate with the bolus drug administration.) . Although these animals would likely be capable of being managed with an optimized dosing schedule, this distinction between Compound 3 and Compound 8 in the long-term cures (>90 days) suggests the prodrug Compound 8 offers significant advantages over the free drug administration. Finally, it is worth noting that these compounds are extraordinarily potent requiring

less than 1 mg of sample to conduct the entire in vivo antitumor testing, suggesting that clinical supplies of such agents could easily be supplied by- chemical synthesis.

Confirming these observations, an analogous antitumor assessment was carried out independently at a second site utilizing a slightly different and harsher protocol for drug administration (neat DMSO vs 30% DMSO in 0.1% glucose) . Although this assessment was terminated after 120 days, it similarly indicates that administration of the prodrug Compound 8 is significantly less toxic than free drug Compound 3, and that it is comparable or superior in terms of reducing deaths due to the disease. Again, 7/10 long-term survivors were observed with Compound 8 at day 120 at the optimal dosing (60 μg/kg) .

A unique class of N-acyl 0-amino phenol prodrugs of CBI-TMI and CBI-indole ₂ was explored as representative members of the duocarmycin and CC-1065 class of antitumor agents. The prodrugs, subject to reductive activation by nucleophilic cleavage of a weak N-O bond, effectively release the free drug in functional cellular assays for cytotoxic activity approaching or matching the activity of the free drug, yet remain essentially stable to ex vivo DNA alkylation conditions (< 0.1-0.01% free drug release), pH 7.0 phosphate buffer, and exhibiting a robust half -life in human plasma ( t^ = 3 hours for Compound 4) . Most impressively, assessment of the in vivo antitumor activity of a representative 0- (acylamino) prodrug, Compound 8, indicate that they approach the potency and exceed the efficacy of the

free drug itself (CBI-indole ₂ ) indicating that the inactive prodrugs not only effectively release the free drug in vivo, but that they offer additional advantages related to a controlled or targeted release in vivo. With cleavage release of the free drug being easily tunable using fundamental chemical principles, the potential of developing derivatives selectively or most effectively released in a reducing hypoxic environment characteristic of solid tumors may further improve on these impressive observations .

Each of the patents, patent applications and articles cited herein is incorporated by reference. The use of the article "a" or "an" is intended to include one or more.

The foregoing description and the examples are intended as illustrative and are not to be taken as limiting. Still other variations within the spirit and scope of this invention are possible and will readily present themselves to those skilled in the art .