AU2003267473A1 - Tumor targeting agents and uses thereof - Google Patents

Description

WO 2004/031218 PCT/F12003/000723 1 TUMOR TARGETING AGENTS AND USES THEREOF FIELD OF THE INVENTION The present invention relates to tumor targeting agents comprising at least one targeting unit and at least one effector unit, as well as to tumor 5 targeting units and motifs. Further, the present invention concerns pharmaceu tical and diagnostic compositions comprising such targeting agents or targeting units, and the use of such targeting agents and targeting units as pharmaceu ticals or as diagnostic tools. The invention further relates to the use of such targeting agents and targeting units for the preparation of pharmaceutical or 10 diagnostic compositions and for the preparation of reagents to be used in di agnosis or research. Furthermore, the invention relates to kits for diagnosing or treating cancer and metastases. Still further, the invention relates to methods of removing, selecting, sorting and enriching cells, and to materials and kits for use in such methods. 15 BACKGROUND OF THE INVENTION Malignant tumors are one of the greatest health problems of man as well as animals, being one of the most common causes of death, also among young individuals. Available methods of treatment of cancer are quite limited, in spite of intensive research efforts during several decades. Although curative 20 treatment (usually surgery in combination with chemothreapy and/or radiother apy) is sometimes possible, malignant tumors (cancer) still are one of the most feared diseases of mankind, requiring a huge number of lives every year. In fact, curative treatment is rarely accomplished if the disease is not diagnosed early. In addition, certain tumor types can rarely, if ever, be treated curatively. 25 There are various reasons for this very undesirable situation but the most important one is clearly the fact that nearly all (if not all) treatment sched ules (except surgery) lack sufficient selectivity. Chemotherapeutic agents commonly used, such as alkylating agents, platinum compounds (e.g. cis platin), bleomycin-type agents, other alkaloids and other cytostatic agents in 3o general, do not act on the malignant cells of the tumors alone but are highly toxic to other cells as well, being usually especially toxic to rapidly dividing cell types, such as hematopoietic and epithelial cells. The same applies to radio therapy. In addition to the above mentioned complications, two further major 35 problems plague the non-surgical treatment of malignant solid tumors. First, WO 2004/031218 PCT/F12003/000723 2 physiological barriers within tumors impede the delivery of therapeutics at ef fective concentrations to all cancer cells. Second, acquired drug resistance re sulting from genetic and epigenetic mechanisms reduces the effectiveness of available drugs. 5 The treatment of cancer patients with currently available, largely non-selective, chemotherapeutic agents or radiotherapy results often also in undesirable side effects. In order to improve the effect of chemotherapeutic agents and to diminish the side effects it would be extremely important to iden tify agents that are capable of targeting to specific organs or tissues or to tu 10 mor tissues and to carry the desired cytotoxic or other drugs specifically to these organs or tissues. The same applies also to a specific field of cancer treatment, namely neutron capture therapy, in which a non-radioactive nucleus (e.g. ' 0 B, 157 Gd or 6 Li) is converted into a radioactive nucleus in vivo in the patient with 15 the aid of thermal (slow) neutrons from an external source. In this case, some prior art agents are claimed to have some 2-3 fold selectivity for at least some types of tumors, but the results obtained have been mainly disappointing and negative. Specific targeting agents would offer remarkable advantages also in this field. 20 Also in the diagnosis of cancer and of metastases, including the fol low-up of patients and the study of the effects of treatment on tumors and me tastases, more reliable, more sensitive and more selective methods and agents would be a great advantage. This is true for all methods currently in use, such as nuclear magnetic resonance imaging (NMR, MRI), X-ray meth 25 ods, histological staining methods (for light microscopy and electron micros copy and related methods, and in the future possibly also NMR, infrared, elec tron spin resonance and related methods) and in general any imaging as well as laboratory methods (histology, cytology, cell sorting, hematological studies, FACS and so on) known by specialists in the field. Here, agents capable of 3o targeting an entity for detection (a spin label, a radioactive substance, a para magnetic contrast agent for NMR or a contrast agent for X-ray imaging or to mography, a boron atom for neutron capture and so on) specifically or selec tively to tumor tissues, metastases or tumor cells and/or to tumor endothelium would be a great advantage. 35 Solid tumor growth is angiogenesis-dependent, and a tumor must continuously stimulate the growth of new microcapillaries for continued growth.

WO 2004/031218 PCT/F12003/000723 3 Tumor blood vessels are structurally and functionally different from their nor mal resting counterparts. In particular, endothelial cells lining new blood ves sels are abnormal in shape, they grow on top of each other and project into the lumen of the vessels. This neovascular heterogeneity depends on the tumor 5 type and on the host organ in which the tumor is growing. Therefore vascular permeability and angiogenesisis are unique in every different organ and in tu mor tissue derived from the organ. There are numerous publications disclosing peptides homing to dif ferent cell and tissue types. Some of these are claimed to be useful as cancer 10 targeting peptides. Among the earliest identified homing peptides described are the integrin and NGR-receptor targeting peptides described by Ruoslahti et al., in e.g., US Patent No 6,180,084. These peptides home to angiogenic vas culature and bind to the NGR-receptor. When tumors switch to the angiogenic phenotype and recruit new 15 blood vessels, endothelial cells in these vessels express proteins on the lu minal surface that are not produced by normal quiescent vascular endothe lium. One such protein is avp3 integrin. US Patent publication, US 6,177,542, discloses a peptide that can bind specifically to axvp3 integrin. The tumor ves sel specific targets described are adhesion molecules that mediate binding of 20 endothelial cells to the vascular basement membrane. This peptide is a nine residue cyclic peptide containing an ArgGlyAsp (RGD) sequence. Pasqualini et al., (1997) showed that when injected intravenously the peptide was able to home to blood vessels of murine and human tumors in mice 40-80 fold more efficiently than to those of control organs. It was suggested that RGD peptides 25 may be suitable tools in tumor targeting for diagnostic and therapeutic pur poses. However, integrin-binding peptides may interfere with cell attachment in general, and are thus not suitable for clinical applications for selective tumor targeting. International Patent Publication WO 00/67771 provides endostatin 3o peptides comprising the amino acid sequence RLQD, RAD, DGK/R. Other ex amples of peptides that home to angiogenic vasculature are described in US Patent Nos 5,817,750 and 5,955,572. These peptides recognize RGD. US Patent 5,628,979 describes oligopeptides for in vivo tumor imag ing and therapy. The oligopeptides contain 4 to 50 amino acids, which contain 35 as a characteristic triplet the amino acid sequence Leu-Asp-Val (LDV). This WO 2004/031218 PCT/F12003/000723 4 triplet is reported to provide the oligopeptide with in vivo binding affinity for LDV binding sites on tumors and other tissues. International Patent publication WO 99/47550 describes cyclic pep tides, containing an HWGF motif, that are specific inhibitors of MMP-2 and 5 MMP-9. They have also found that the cyclic decapeptide CTTHWGFTLC spe cifically inhibits the activities of these enzymes, suppresses migration of both tumor cells and endothelial cells in vitro, homes to tumor vasculature in vivo, and prevents the growth and invasion of tumors in mice. However, peptides that act as inhibitors of MMPs show background binding to non-tumor tissues. 10 The fact that MMPs are expressed also in normal tissue throughout the body also makes the administration of such peptides to humans or animals hazard ous and even fatal, since the activity of these enzymes is required for normal tissue functions (Hidalgo and Eckhardt, 2001). US Patent publication US 2002/0102265A1 describes a peptide, 15 TSPLNIHNGQKL, that targets squamous cell cancer cell lines, and becomes internalized into cells in vitro. This peptide also targets experimental squamous carcinomas in nude mice. US Patent Nos. 5,622,699 and 6,068,829 disclose a family of pep tides comprising an SRL motif, which selectively home to brain. 20 International Patent publication WO 02/20769 discloses methods for identifying tissue specific peptides by phage display and biopanning. Some of the identified peptides are suggested to be tumor specific. Although there are known homing peptides that bind to tumor vas culature, there are still very scarce reports on targeting agents that actually 25 target tumor cells and tissues in vivo. Most of the previously described target ing peptides are vasculature specific. Thus, there is still an established need for new agents that target selectively to tumor tissue, tumor vasculature, or both. For therapeutic applications, targeting peptides have been conju 30 gated to doxorubicin in an uncontrolled fashion, obviously resulting in mixtures of products or at least in an undefined structure and possibly also resulting in unefficient action and especially in difficulties in the identification, purification, quality control and quantitative analysis of the agent, even the amount of doxorubicin per peptide molecule remaining unknown (e.g. Arap et al., 1998). 35 The unspecific conjugation process might also impair the targeting functions of the peptide.

WO 2004/031218 PCT/F12003/000723 5 Another very serious disadvantage of the prior art is that most of the described targeting peptides appear to target to the tumor endothelium only and not to the tumor mass itself. For example, the targeting peptide used by Nicklin et al. (2000) directed adenovirus DNA transfection to resting endothelial 5 cells in vitro, under conditions that hardly could be applied in vivo. The targeting units according to the present invention offer an ad vantage over the prior art in that they seem to target to both the tumor endo thelium and the tumor cell mass. This fact provides the possibility to target and destroy tumor endothelium supporting tumor growth as well as the tumor mass 10 itself. A major advantage of this approach comes from the fact that the endo thelium is a genetically stable tissue that will not acquire drug resistance but will be irreversibly eliminated. It is not known whether the prior art targeting peptides are universal in the sense of being capable to target to any malignant tumor type. Thus, their 15 use as targeting therapeutic agents to a certain specified tumor may be com pletely useless, giving no therapeutic advantage or effect over the free thera peutic agent itself. An even more serious drawback is that the use of such tar geting agents in diagnostic procedures may not reveal all existing tumors and the malignant process may remain unrecognized. 20 The present invention offers a significant improvement in view of the prior art, since the targeting agents here described were found to target to all of the various tumor types tested. Remarkably, they target, for example, sar comas, such as Kaposi's sarcoma, ornithine decarboxylase (ODC) overex pressing, highly angiogenic tumors, carcinomas, and to human primary and 25 metastatic melanomas Brief Description of the Invention It is an object of the present invention to provide novel tumor and angiogenic tissue targeting agents that comprise at least one targeting unit and, optionally, at least one effector unit. In particular, the invention provides 30 targeting units comprising at least one motif that is capable of targeting both tumor endothelium and tumor cell mass. Such targeting units, optionally cou pled to at least one effector unit, are therapeutically and diagnostically useful, especially in the treatment and diagnosis of cancer, including metastases. Fur thermore the targeting agents according to the present invention are useful for 35 cell removal, selection, sorting and enrichment.

WO 2004/031218 PCT/F12003/000723 6 It is a second object of this invention to provide pharmaceutical and diagnostic compositions comprising at least one targeting agent or at least one targeting unit comprising at least one motif capable of specifically targeting tu mors, tumor cells and tumor endothelium. 5 Further, it is a third object of the invention to provide novel diagnos tic and therapeutic methods and kits for the treatment and/or diagnosis of can cer. The present invention is based on the finding that a group of pep tides having specific amino acid sequences or motifs are capable of selectively 10 targeting tumors in vivo and tumor cells in vitro. Thus, the peptides of this in vention, when administered to a human or animal subject, are capable of se lectively binding to tumors but not to normal tissue in the body. The present invention is also directed to the use of the targeting agents and analogues thereof for the manufacture of a pharmaceutical or 15 diagnostic composition for treating or diagnosing cancer. The targeting units of this invention may be used as such or coupled to at least one effector unit. Such substances can destroy the tumors or hinder their growth. The targeting units and targeting agents of this invention can tar get also metastases and therefore they may be used to destroy or hinder the 20 growth of metastases. As early diagnosis of metastases is very important for successful treatment of cancer, an important use of the targeting units and tar geting agents of this invention is in early diagnosis of tumor metastases. The present invention further encompasses salts, derivatives and analogues of the targeting units and targeting agents, as described herein, as 25 well as uses thereof. It is a further object of the present invention to provide diagnostic and pharmaceutical compositions comprising targeting agents according to the present invention, as well as therapeutic and diagnostic methods for the treat ment and diagnosis of cancer, utilizing targeting agents according to the pre 30 sent invention. Also provided are kits for use in such methods or for research purposes, as well as in cell sorting or removal. Especially preferred embodiments of the present invention relate to a group of small, cyclic tumor targeting peptides comprising a motif, LRS or SRL, optionally coupled to an effector unit and other additional units, as de 35 scribed in more detail herein.

WO 2004/031218 PCT/F12003/000723 7 BRIEF DESCRIPTION OF THE DRAWINGS Figure 1 is a graph showing the therapeutic effect of a targeting agent comprising doxorubicin. DETAILED DESCRIPTION OF THE INVENTION 5 For the purpose of this invention, the term "cancer" is used herein in its broadest sense, and includes any disease or condition involving trans formed or malignant cells. In the art, cancers are classified into five major cate gories, according to their tissue origin (histological type): carcinomas, sarco mas, myleomas, and lymphomas, which are solid tumor type cancers, and leu 10 kemias, which are "liquid cancers". The term cancer, as used in the present in vention, is intended to primarily include all types of diseases characterized by solid tumors, including disease states where there is no detectable solid tumor or where malignant or transformed cells, "cancer cells", appear as diffuse infil trates or sporadically among other cells in healthy tissue. 15 The terms "amino acid"and "amino alcohol" are to be interpreted herein to include also diamino, triamino, oligoamino and polyamino acids and alcohols; dicarboxyl, tricarboxyl, oligocarboxyl and polycarboxyl amino acids; dihydroxyl, trihydroxyl, oligohydroxyl and polyhydroxyl amino alcohols; and analogous compounds comprising more than one carboxyl group or hydroxyl 20 group and one or more amino groups. By the term "peptide" is meant, according to established terminol ogy, a chain of amino acids (peptide units) linked together by peptide bonds to form an amino acid chain. Peptides may be cyclic as described below. For the purposes of the present invention, also compounds comprising one or more D 25 amino acids, p-amino acids and/or other unnatural amino acids (e.g. amino ac ids with unnatural side chains) are included in the term "peptide". For the pur poses of the present invention, the term "peptide" is intended to include pepti dyl analogues comprising modified amino acids. Such modifications may com prise the introduction or presence of a substituent in a ring or chain; the intro so duction or presence of an "extra" functional group such as an amino, hy drazino, carboxyl, formyl (aldehyde) or keto group, or another moiety; and the absence or removal of a functional group or other moiety. The term also in cludes analogues modified in the amino- and/or carboxy termini, such as pep tide amides and N-substituted amides, peptide hydrazides, N-substituted hy 35 drazides, peptide esters, and their like, and peptides that do not comprise the WO 2004/031218 PCT/F12003/000723 8 amino-terminal -NH 2 group or that comprise e.g. a modified amino-terminal amino group or an imino or a hydrazino group instead of the amino-terminal amino group, and peptides that do not comprise the carboxy-terminal carboxyl group or comprise a modified group instead of it, and so on. 5 Some examples of possible reaction types that can be used to mod ify peptides, forming "peptidyl analogues", are e.g., cycloaddition, condensa tion and nucleophilic addition reactions as well as esterification, amide forma tion, formation of substituted amides, N-alkylation, formation of hydrazides, salt formation. Salt formation may be the formation of any type of salt, such as al 10 kali or other metal salt, ammonium salt, salts with organic bases, acid addition salts etc. Peptidyl analogues may be synthesized either from the correspond ing peptides or directly (via other routes). Compounds that are structural or functional analogues of the pep tides of the invention may be compounds that do not consist of amino acids or 15 not of amino acids alone, or some or all of whose building blocks are modified amino acids. Different types of building blocks can be used for this purpose, as is well appreciated by those skilled in the art. The function of these compounds in biological systems is essentially similar to the function of the peptides. The resemblance between these compounds and the original peptides is thus 20 based on structural and functional similarities. Such compounds are called peptidomimetic analogues, as they mimic the function, conformation and/or structure of the original peptides and, for the purposes of the present invention, they are included in the term "peptide". A functional analog of a peptide according to the present invention 25 is characterized by a binding ability with respect to the binding to tumors, tumor tissue, tumor cells or tumor endothelium which is essentially similar to that of the peptides they resemble. For example, compounds like benzolactam or piperazine containing analogues based on the primary sequence of the original peptides can be 30 used (Adams et al., 1999; Nakanishi and Kahn, 1996a, 1996b; Houghten et al., 1999; Nargund et al., 1998). A large variety of types of peptidomimetic sub stances have been reported in the scientific and patent literature and are well known to those skilled in the art. Peptidomimetic substances (analogues) may comprise for example one or more of the following structural components: re 35 duced amides, hydroxyethylene and/or hydroxyethylamine isosteres, N-methyl amino acids, urea derivatives, thiourea derivatives, cyclic urea and/or thiourea WO 2004/031218 PCT/F12003/000723 9 derivatives, poly(ester imide)s, polyesters, esters, guanidine derivatives, cyclic guanidines, imidazoyl compounds, imidazolinyl compounds, imidazolidinyl compounds, lactams, lactones, aromatic rings, bicyclic systems, hydantoins and/or thiohydantoins as well as various other structures. Many types of com 5 pounds for the synthesis of peptidomimetic substances are available from a number of commercial sources (e.g. Peptide and Peptidomimetic Synthesis, Reagents for Drug Discovery, Fluka ChemieGmbH, Buchs, Switzerland, 2000 and Novabiochem 2000 Catalog, Calbiochem-Novabiochem AG, Laufelfingen, Switzerland, 2000). The resemblance between the peptidomimetic compounds 10 and the original peptides is based on structural and/or functional similarities. Thus, the peptidomimetic compounds mimic the properties of the original pep tides and, for the purpose of the present application, their binding ability is similar to the peptides that they resemble. Peptidomimetic compounds can be made up, for example, of unnatural amino acids (such as D-amino acids or 15 amino acids comprising unnatural side chains, or of p-amino acids etc.), which do not appear in the original peptides, or they can be considered to consist of or can be made from other compounds or structural units. Examples of syn thetic peptidomimetic compounds comprise N-alkylamino cyclic urea, thiourea, polyesters, poly(ester imide)s, bicyclic guanidines, hydantoins, thiohydantoins, 20 and imidazol-pyridino-inoles (Houghten et al. 1999 and Nargund et al., 1998). Such peptidomimetic compounds can be characterized as being "structural or functional analogues" of the peptides of this invention. For the purpose of the present invention, the term "targeting unit" stands for a compound, a peptide, capable of selectively targeting and selec 25 tively binding to tumors, and, preferably, also to tumor stroma, tumor paren chyma and/or extracellular matrix of tumors. Another term used in the art for this specific association is "homing". Tumor targeting means that the targeting units specifically bind to tumors when administered to a human or animal body. More specifically, the targeting units may bind to a cell surface, to a specific 30 molecule or structure on a cell surface or within the cells, or they may associ ate with the extracellular matrix present between the cells. The targeting units may also bind to the endothelial cells or the extracellular matrix of tumor vas culature. The targeting units may bind also to the tumor mass, tumor cells and extracellular matrix of metastases. 35 Generally, the terms "targeting" or "binding" stand for adhesion, at tachment, affinity or binding of the targeting units of this invention to tumors, WO 2004/031218 PCT/F12003/000723 10 tumor cells and/or tumor tissue to the extent that the binding can be objectively measured and determined e.g., by peptide competition experiments in vivo or ex vivo, on tumor biopsies in vitro or by immunological stainings in situ, or by other methods known by those skilled in the art The exact mechanism of the 5 binding of targeting units according to the present invention is not known. Tageting peptides according to the present invention are considered to be "bound" to the tumor target in vitro, when the binding is strong enough to with stand normal sample treatment, such as washes and rinses with physiological saline or other physiologically acceptable salt or buffer solutions at physiologi 10 cal pH, or when bound to a tumor target in vivo long enough for the effector unit to exhibit its function on the target. The binding of the present targeting agents or targeting units to tu mors is "selective" meaning that they do not bind to normal cells and organs, or bind to such to a significantly lower degree as compared to tumor cells and 15 organs. Pharmaceutically and diagnostically acceptable salts of the target ing units and agents of the present invention include salts, esters, amides, hy drazides, N-substituted amides, N-substituted hydrazides, hydroxamic acid de rivatives, decarboxylated and N-substituted derivatives thereof. Suitable phar 20 maceutically acceptable salts are readily acknowledged by those skilled in the art. TARGETING MOTIFS ACCORDING TO THE PRESENT INVENTION It has now surprisingly been found that a three-amino-acid motif Dd Ee-Ff, wherein Dd-Ee-Ff is either Aa-Bb-Cc or Cc-Bb-Aa, and 25 Aa is isoleucine, leucine or tert-leucine, or a structural or functional analogue thereof; Bb is arginine, homoarginine or canavanine, or a structural or functional ana logue thereof; and Cc is serine or homoserine, or a structural or functional analogue thereof, 30 targets and exhibits selective binding to tumors and cancers and tumor cells and cancer cells. Aa according to the present invention may comprise in its sidechain a branched, non-branched or alicyclic structure with at least two siminal or dif ferent atoms selected from the group consisting of carbon, silicon, halogen 35 bonded to carbon, ether-oxygens and thioether-sulphur. The analogue may be selected from the group consisting of branched, non-branched or cyclic WO 2004/031218 PCT/F12003/000723 11 non-aromatic, lipophilic and hydrophobic amino acids or amino acid analogues or derivatives or structural and/or functional analogues thereof; amino acids or carboxylic acids or amino acid analogues or derivatives or carboxylic acid ana logues or derivatives having one or more lipophilic carborane-type or other 5 lipophilic boron-containing side chains or other lipophilic cage-type structures. Aa may be selected from the group consisting of: 1) a-amino acids whose side chain is one of the following: - ethyl - propyl 10 - 1-methylpropyl (the side chain of isoleucine) - 2-methylpropyl (the side chain of leucine) - 2,2-dimethylpropyl - 1 -ethylpropyl - tert-butyl 15 - tert-pentyl - 3-methylbutyl - 2-methylbutyl - methylbutyl - ethylbutyl 20 - 2-ethylbutyl - cyclohexyl - 2-methylcyclohexyl - cyclopentyl - 2-methylcyclopentyl 25 - 3-methylcyclohexyl - cyclobutyl - cyclopropyl - 2-methylcyclopropyl - methoxyethyl 30 - methoxyethyl - methoxymethyl - ethoxymethyl - 2-ethoxyethyl - 1 -ethoxyethyl 35 - 2-methoxypropyl - 2,2-dimethoxypropyl WO 2004/031218 PCT/F12003/000723 12 - 1 -methylpropyl - 1 -methylbutyl - 1 -methylpentyl - 1,1-dimethylpropyl 5 - 1,1-dimethylbutyl - 1,1-dimethylpentyl - 1,2-dimethylpropyl - 1 -cyclopropylethyl - 2-cyclopropylethyl 10 - cyclopropylmethyl - 1 -cyclopropylethyl - 1 -cyclopropylpropyl - 2-cyclopropylpropyl - 3-cyclopropyipropyl 15 - any cyclobutylalkyl - 1 -ethylpropyl - 1 -methylethyl - other mono-, di-, tri- or oligoalkyl-alkyl - other cyclic alkyl or substituted cyclic alkyl or alkyl that is substituted 20 with one or more substituted or unsubstituted cycloalkyl group(s) and optionally one or more alkyl group(s) - allyl - vinyl - 1 -methylallyl 25 - 1 -ethylallyl - 1 -ethylvinyl - 1-propenyl - 1-methyl-1-propenyl - methyl-1-propenyl 30 - methyl-1-propenyl - 1 -ethyl-1 -propenyl - ethyl-1 -propenyl - ethyl-1 -propenyl - 1 -methyl-1 -butenyl 35 - methyl-1-butenyl - methyl-1-butenyl WO 2004/031218 PCT/F12003/000723 13 - 1-ethyl-1 -butenyl - 2- ethyl-1 -butenyl - ethyl-2-butenyl - ethyl-2-butenyl 5 - ethyl-3-butenyl - ethyl-3-butenyl - ethyl-3-butenyl 2) any of the following carboxylic acids, including any optical isomers thereof - 4-methylpentanoic acid 10 - 3- methylpentanoic acid - 4,4-dimethylpentanoic acid - 3,4-dimethylpentanoic acid - 3,3-dimethylpentanoic acid - 3-methylhexanoic acid 15 - 4-methylhexanoic acid - 5-methylhexanoic acid - 2-ethylpentanoic acid - 3-ethylpentanoic acid - 4-ethylpentanoic acid 20 - 2-cyclopropylpentanoic acid - 3-cyclopropylpentanoic acid - 4-cyclopropylpentanoic acid - 2-methylbutanoic acid - 3-methylbutanoic acid 25 - 4-methylbutanoic acid - 2-cyclopropylbutanoic acid - 3-cyclopropylbutanoic acid - 4-cyclopropylbutanoic acid 3) any optical and geometrical isomer of any of the following compounds: 30 - 2-amino-4-methyl-3-pentenoic acid - 2-amino-4-methyl-4-pentenoic acid - 2-amino-5-methyl-3-hexenoic acid - 2-amino-5-methyl-4-hexenoic acid - 2-amino-5-methyl-5-hexenoic acid 35 and WO 2004/031218 PCT/F12003/000723 14 4) aminosubstituted (N-substituted) analogues of the amino-comprising com pounds of points 1 and 3 that bear at the amino group - one methyl, ethyl, propyl, isopropyl or other alkyl group - one cycloalkyl group 5 - one 9-fluorenylmethyloxycarbonyl (FMOC) group - one benzyloxycarbonyl (Cbz) group - one tert-butyloxycarbonyl (BOC) group - two identical, similar and/or different groups selected from the ones men tioned above in this point (point 4). 10 Aa may also be an a-amino acid (either L- or D- amino acid) of the formula R' - CR 2

(NH

2 ) - COOH wherein the side chain R 1 is selected from the side chains listed above, and the side chains R 2 is selected from the group consisting of: hydrogen, methyl, ethyl, propyl. Bb according to the present invention may be selected from the 15 group consisting of amino acids or structural or functional analogues thereof containing one or more guanyl groups, aminodino groups or their analogues and derivatives and structural or functional equivalents; one or more groups containing at least two nitrogen atoms each and have or can gain a delocal ized positive charge. 20 Bb may be selected from the group of compounds of the following formula: 2 3 R R

N--R

5

(CH

2 )n H-C-COOH

NH

2 wherein R1 - R5 is hydrogen or methyl, R2 and R3 may form -CH2-CH2- and 25 n is 1-6.

WO 2004/031218 PCT/F12003/000723 15 Preferably, Bb is the L- or D- form of arginine, homoarginine, canavanine, 5 2-amino-8-guanidino-octanoic acid, 2-amino-7-guanidino-octanoic acid, 2-amino-6-guanidino-octanoic acid, 2-amino-5-guanidino-octanoic acid, 2-amino-7-guanidino-heptanoic acid, 10 2-amino-6-guanidino-heptanoic acid, 2-amino-5-guanidino-heptanoic acid, 2-amino-4-guanidino-heptanoic acid, 2-amino-5-guanidino-hexanoic acid, 2-amino-4-guanidino-hexanoic acid, 15 2-amino-3-guanidino-hexanoic acid, 2-amino-4-guanidino-pentanoic acid, 2-amino-3-guanidino-pentanoic acid. Cc according to the present invention may be selected from the group consisting of amino acids, amino alcohols, diamino alcohols, tri-, oligo 20 and polyamino alcohols and amino acid analogues, derivatives and structural or functional analogues thereof, comprising one or more hydroxyl group(s), es terified hydroxyl group(s), methoxyl group(s) and/or other etherified hydroxyl (ether) groups. Cc as defined above may be serine or homoserine or a structural or 25 functional analogue thereof, comprising at least one hydroxyl group; or may be selected from the group consisting of: any other monoaminocarboxylic acid comprising at least one alcoholic hydroxyl group any carboxylic acid comprising at least one alcoholic hydroxyl group 30 any other aminocarboxylic acid comprising an aliphatic or other side chain that comprises one or more alcoholic hydroxyl (OH) function(s) and/or esterified hydroxyl function(s). Preferably, Cc is the L- or D- form of serine, 35 homoserine, 2-amino-7-hydroxyheptanoic acid, WO 2004/031218 PCT/F12003/000723 16 2-amino-5-hydroxypentanoic acid, 2-amino-6-hydroxyhexanoic acid, 2-amino-8-hydroxyoctanoic acid, or any other hydroxy-2-aminocarboxylic acid. 5 Alternatively, the motif Aa-Bb-Cc, as a whole, according to the pre sent invention is a structural or functional analogue of a structure where Aa, Bb and Cc are as defined above. Preferred embodiments of the present invention include tumor tar geting motifs Aa-Bb-Cc selected from those given in Table 1 as well as struc 10 tural and functional analogues thereof. TABLE 1 Aa Bb Cc 1 L-isoleucine L-arginine L-serine 2 " " L-homoserine 3 D-isoleucine D-arginine D-serine 4 " " D-homoserine 5 L-leucine L-arginine L-serine 6 " " L-homoserine 7 D-leucine D-arginine D-serine 8 "" D-homoserine 9 L-isoleucine L-homoarginine L-serine 10 " " L-homoserine 11 D-isoleucine D-homoarginine D-serine 12 ' " D-homoserine 13 L-leucine L-homoarginine L-serine 14 L-homoserine 15 D-leucine D-homoarginine D-serine 16 " " D-homoserine 17 L-2-aminopentanoic acid L-arginine L-serine 18 D-2-aminopentanoic acid D-arginine D-serine 19 L-2-aminopentanoic acid L-arginine L-homoserine 20 D-2-aminopentanoic acid D-arginine D-homoserine 21 L-2-aminohexanoic acid L-arginine L-serine 22 D-2-aminohexanoic acid D-arginine D-serine 23 L-2-aminohexanoic acid L-arginine L-homoserine WO 2004/031218 PCT/F12003/000723 17 24 D-2-aminohexanoic acid D-arginine D-homoserine 25 L-2-aminoheptanoic acid L-arginine L-serine 26 D-2-aminoheptanoic acid D-arginine D-serine 27 L-2-aminoheptanoic acid L-arginine L-homoserine 28 D-2-aminoheptanoic acid D-arginine D-homoserine 29 L--2-amino-2-ethylbutanoic acid L-arginine L-serine 30 D-2-amino-2-ethylbutanoic acid D-arginine D-serine 31 L-2-amino-2-ethylbutanoic acid L-arginine L-homoserine 32 D-2-amino-2-ethylbutanoic acid D-arginine D-homoserine 33 L-isoleucine L-arginine 2-amino-7-hydroxyheptanoic acid 34 D-isoleucine D-arginine 2-amino-7-hydroxyheptanoic acid 35 L-leucine D-arginine 2-amino-7-hydraxyheptanoic acid 36 D-leucine D-arginine 2-amino-7-hydroxyheptanoic acid 37 L-isoleucine L-arginine L-2-amina-5-hydroxypentanoic acid 38 D-isoleucine D-arginine D-2-amino-5-hydroxypentanoic acid 39 L-leucine L-arginine L-2-amino-5-hydroxypentanoic acid 40 D-leucine D-arginine D-2-amino-5-hydroxypentanoic acid 41 L-isoleucine L-arginine L-2-amino-6-hydroxyhexanoic acid 42 D-isoleucine D-arginine D-2-amino-6-hydroxyhexanoic acid 43 L-leucine L-arginine L-2-amino-6-hydroxyhexanoic acid 44 D-leucine D-arginine D-2-amino-6-hydroxyhexanoic acid 45 L-2-aminopentancic acid L-homoarginine L-serine 46 D-2-aminopentanoic acid D-homoarginine D-serine 47 L-2-aminopentanoic acid L-homoarginine L-homoserine 48 D-2-aminopentanoic acid D-homoarginine D-homoserine 49 L-2-aminohexanoic acid L-homoarginine L-serine 50 D-2-aminohexanoic acid D-homoarginine D-serine 51 L-2-aminohexanoic acid L-homoarginine L-homoserine 52 D-2-aminohexanoic acid D-homoarginine D-homoserine 53 L-2-aminoheptanoic acid L-homoarginine L-serine 54 D-2-aminoheptanoic acid D-homoarginine D-serine 55 L-2-aminoheptanoic acid L-homoarginine L-homoserine 56 D-2-aminoheptanoic acid D-homcarginine D-homoserine 57 L-2-amino-2-ethylbutanoic acid L-homoarginine L-serine 58 D-2-amino-2-ethylbutanoic acid D-homoarginine D-seririe 59 L-2-amino-2-ethylbutanoic acid L-homoarginine L-homoserine WO 2004/031218 PCT/F12003/000723 18 60 D-2-amino-2-ethylbutanoic acid D-homoarginine D-homoserine 61 L-isoleucine L-homoarginine 2-amino-7-hydroxyheptanoic acid 62 D-isoleucine D-homoarginine 2-amino-7-hydroxyheptanoic acid 63 L-leucine D-homoarginine 2-amino-7-hydroxyheptanoic acid 64 D-leucine D-homoarginine 2-amino-7-hydroxyheptanoic acid 65 L-isoleucine L-homoarginine L-2-amino-5-hydroxypentanoic acid 66 D-isoleucine D-homoarginine D-2-amino-5-hydroxypentanoic acid 67 L-leucine L-homoarginine L-2-amino-5-hydroxypentanoic acid 68 D-leucine D-homoarginine D-2-amino-5-hydroxypentanoic acid 69 L-isoleucine L-homoarginine L-2-amino-6-hydroxyhexanolc acid 70 D-isoleucine D-homoarginine D-2-amino-6-hydroxyhexanoic acid 71 L-Ieucine L-homoarginine L-2-amino-6-hydroxyhexanoic acid 72 D-leucine D-homoarginine D-2-amino-6-hydroxyhexanoic acid Thus, typical and preferred characteristics of Aa include lipofilicity, hydrophobicity and aliphatic character in at least one side chain, wheras Bb in cludes a delocalized positive charge and Cc has the ability of participating in 5 OH-binding. Especially preferred motifs Dd-Ee-Ff according to the present inven tion are leucine-arginine-serine (LRS) and serine-arginine-leucine (SRL). The motifs Dd-Ee-Ff according to the present invention may form part of a larger structure, such as a peptide or some other structure. When the 10 compound or structure in question comprises more than one motif Dd-Ee-Ff, the orientation and direction of the motifs may vary. TARGETING UNITS ACCORDING TO THE PRESENT INVENTION It has also been found that peptides and structural or functional ana logues thereof comprising a tumor targeting motif according to the present in 15 vention target to and exhibit selective binding to tumor cells and tissues. Pep tides comprising a tumor targeting motif according to the present invention and, optionally, up to four additional amino acid residues or analogues thereof, likewise exhibit such targeting and selective binding and are especially pre ferred embodiments of the present invention. 20 Such peptides are highly advantageous for use as targeting units according to the present invention, e.g., because of their small size and their WO 2004/031218 PCT/F12003/000723 19 easier and more reliable and much cheaper synthesis, purification, analysis and quality control. Especially preferred targeting units according to the present inven tion are peptides cyclized by an amide bond, such as a lactam bridge or by an 5 ester bond, such as a lacotone bridge. Such cyclic targeting units have now been shown to selectively tar get to tumors in vivo. Preferred tumor targeting units according to the present invention comprise a tumor targeting motif Dd-Ee-Ff as defined above, and additional 10 residues selected from the group consisting of: natural amino acids; unnatural amino acids; amino acid analogues comprising maximally 30 non-hydrogen atoms and an unlimited number of hydrogen atoms,; and 15 other structural units and residues whose molecular weight and/or formula weight is maximally 270; wherein the number of said additional residues ranges from 0 to 4, preferably from 0 to 3, more preferably from 0 to 2, and most preferably is either 0 or 2. 20 Cyclic peptides are usually more stable in vivo and in many other biological systems than are their non-cyclic counterparts, as is known in the art. It has now, however, surprisingly been found that the targeting property of the small peptides according to the present invention is more pronounced when the targeting unit is cyclic or contained in a cyclic structure. 25 Preferred targeting units according to the present invention may comprise a structure Cy-Rrn-Dd-Ee-Ff-Rrm-Cyy 30 wherein, Dd-Ee-Ff is a tumor targeting motif according to the present inven tion, as described above, Rr is any amino acid residue, n and m are integers 0-4, preferably 0-3, more preferably 0-2, 35 whereby the sum of n and m does not exceed four, and, WO 2004/031218 PCT/F12003/000723 20 Cy and Cyy are entities capable of forming a cyclic structure by means of a lactam ring. Lactams can be of several subtypes, such as "head to tail" (carboxy terminus plus amino terminus), "head to side chain" and "side chain to head" 5 (carboxy or amino terminus plus one side chain amino or carboxyl group) and "side chain to side chain" (amino groug of one side chain and carboxys group of another side chaine). Preferred targeting units are such, where Rr is any amino acid resi due, except histidine, lysine or tryptophane. Especially preferred are targeting 10 units wherein Rr is R or G. Preferred structures are thus compounds of the general formula Cy-Rrn-Dd-Ee-Ff- Rrm-Cyy 15 as defined above, and wherein Cy and Cyy are residues capable of forming a lactam bond, such as aspartic acid (D), glutamic acid (E), lysin (K), ornithine (0) or analogues thereof comprising no more than 12 carbon atoms. Especially preferred targeting units according to the present inven tion having a cyclic structure by virtue of a lactam bridge, are : 20 DLRSK (SEQ ID NO: 1), DGRGLRSK (SEQ ID NO: 2), DRGLRSK (SEQ ID NO: 3), DRYYNLRSK (SEQ ID NO: 4), DSRYNLRSK (SEQ ID NO: 5), DLRSGRK (SEQ ID NO: 6), DLRSGRGK (SEQ ID NO: 7), OLRSE (SEQ ID NO. 8), OLRSGRGE (SEQ ID NO. 9) and KLRSD (SEQ ID NO. 10), as well as salts, esters, derivatives and analogues thereof, as defined above. 25 TARGETING AGENTS ACCORDING TO THE PRESENT INVENTION It has also been found that targeting agents comprising at least one tumor targeting unit according to the present invention, and at least one effec tor unit, target to and exhibit selective binding to cancer cells and tissues as well as endothelial cells. 30 The tumor targeting agents according to the present invention may optionally comprise unit(s) such as linkers, solubility modifiers, stabilizers, charge modifiers, spacers, lysis or reaction or reactivity modifiers, internalizing units or internalization enhancers or membrane interaction units or other local route, attachment, binding and distribution affecting units. Such additional units 35 of the tumor targeting agents according to the present invention may be cou pled to each other by any means suitable for that purpose.

WO 2004/031218 PCT/F12003/000723 21 Many possibilities are known to those skilled in the art for linking structures, molecules, groups etc. of the types in question or of related types, to each other. The various units may be linked either directly or with the aid of one or more identical, similar and/or different linker units. The tumor targeting 5 agents of the invention may have different structures such as any of the non limiting types schematically shown below: 1. EU - TU 2. (EU)a - (TU)m 10 3. (EU)n - (TU)m - (EU)k 4. TU EU 15 TU 5. EU TU 20 EU where EU indicates "effector unit" and TU indicates "targeting unit" and n, m and k are independently any integers except 0. In the targeting agents according to the present invention, as in 25 many other medicinal and other substances, it may be wise to include spacers or linkers, such as amino acids and their analogues, such as long-chain omega-amino acids, to prevent the targeting units from being 'disturbed', steri cally, electronically or otherwise hindered or 'hidden' by effector units or other unit of the targeting agent. 30 In targeting agents according to the present invention it may be use ful for increased activity to use dendrimeric or cyclic structures to provide a possiblility to incorporate multiple effector units or additional units per targeting unit. Preferred targeting agents according to the present invention com 35 prise a structure WO 2004/031218 PCT/F12003/000723 22 Ef-TU-Eff, wherein TU is a targeting unit according to the present invention as defined abover; and 5 Ef and Eff are selected from the group consisting of: effector units, linker units, solubility modifier units, stabilizer units, charge modi fier units, spacer units, lysis and/or reaction and/or reactivity modifier units, in ternalizing and/or internalization enhancer and/or membrane interaction units and/or other local route and/or local attachment/local binding and/or distribu 10 tion affecting units, adsorption enhancer units, and other related units; and peptide sequences and other structures comprising at least one such unit; and peptide sequences comprising no more than 20, preferably no more than 12, more preferably no more than 6, natural and/or unnatural amino acids; and natural and unnatural amino acids comprising no more than 25 non-hydrogen 15 atoms and an unlimited number of hydrogen atoms; as well as salts, esters, derivatives and analogues thereof. EFFECTOR UNITS For the purposes of this invention, the term "effector unit" means a molecule or radical or other chemical entity as well as large particles such as 20 colloidal particles and their like; liposomes or microgranules. Suitable effector units may also consitute nanodevices or nanochips or their like; or a combina tion of any of these, and optionally chemical structures for the attachment of the constituents of the effector unit to each or to parts of the targeting agents. Effector units may also contain moieties that effect stabilization or solubility 25 enhancement of the effector unit. Preferred effects provided by the effector units according to the pre sent invention are therapeutical (biological, chemical or physical) effects on the targeted tumor; properties that enable the detection or imaging of tumors or tumor cells for diagnostic purposes; or binding abilities that relate to the use of 30 the targeting agents in different applications. A preferred (biological) activity of the effector units according to the present invention is a therapeutic effect. Examples of such therapeutic activi ties are for example, cytotoxicity, cytostatic effect, ability to cause differentia tion of cells or to increase their degree of differentiation or to cause phenotypic 35 changes or metabolic changes, chemotactic activities, immunomodulating ac tivities, pain relieving activities, radioactivity, ability to affect the cell cycle, abil- WO 2004/031218 PCT/F12003/000723 23 ity to cause apoptosis, hormonal activities, enzymatic activities, ability to trans fect cells, gene transferring activities, ability to mediate "knock-out" of one or more genes, ability to cause gene replacements or "knock-in", antiangiogenic activities, ability to collect heat or other energy from external radiation or elec 5 tric or magnetic fields, ability to affect transcription, translation or replication of the cell's genetic information or external related information; and to affect post-transcriptional and/or post-translational events. Other preferred therapeutic approaches enabled by the effector units according to the present invention may be based on the use of thermal 10 (slow) neutrons (to make suitable nuclei radioactive by neutron capture), or the administration of an enzyme capable of hydrolyzing for example an ester bond or other bonds or the administration of a targeted enzyme according to the present invention. Examples of preferred functions of the effector units according to 15 the present invention suitable for detection are radioactivity, paramagnetism, ferromagnetism, ferrimagnetism, or any type of magnetism, or ability to be de tected by NMR spectroscopy, or ability to be detected by EPR (ESR) spectros copy, or suitability for PET and/or SPECT imaging, or the presence of an im munogenic structure, or the presence of an antibody or antibody fragment or 20 antibody-type structure, or the presence of a gold particle, or the presence of biotin or avidin or other protein, and/or luminescent and/or fluorescent and/or phosphorescent activity or the ability to enhance detection of tumors, tumor cells, endothelial cells and metastases in electron microscopy, light microscopy (UV and/or visible light), infrared microscopy, atomic force microscopy or tun 25 neling microscopy, and so on. Preferred binding abilities of an effector unit according to the pre sent invention include, for example: a) ability to bind to a substance or structure such as a histidine or other tag, b) ability to bind to biotin or analogues thereof, 30 c) ability to bind to avidin or analogues thereof, d) ability to bind to an enzyme or a modified enzyme, e) ability to bind metal ion(s) e.g. by chelation, f) ability to bind a cytotoxic, apoptotic or metobolism affectin substance or a substance capable of being converted in situ into such a substance, 35 g) ability to bind to integrins and other substances involved in cell adhesion, migration, or intracellular signaling, WO 2004/031218 PCT/F12003/000723 24 h) ability to bind to phages, i) ability to bind to lymphocytes or other blood cells, j) ability to bind to any preselected material by virtue of the presence of antibodies or structures selected by biopanning, 5 k) ability to bind to material used for signal production or amplification, I) ability to bind to therapeutic substance. Such binding may be the result of e.g. chelation, formation of cova lent bonds, antibody-antigen-type affinity, ion pair or ion associate formation, specific interactions of the avidin-biotin-type, or the result of any type or mode 10 of binding or affinity. One or more of the effector units or parts of them may also be a part of the targeting units themselves. Thus, the effector unit may for example be one or more atoms or nuclei of the targeting unit, such as radioactive atoms or atoms that can be made radioactive, or paramagnetic atoms or atoms that are 15 easily detected by MRI or NMR spectroscopy (such as carbon-13). Further ex amples are, for example, boron-comprising structures such as carborane-type lipophilic side chains. The effector units may be linked to the targeting units by any type of bond or structure or any combinations of them that are strong enough so that 20 most, or preferably all or essentially all of the effector units of the targeting agents remain linked to the targeting units during the essential (necessary) targeting process, e.g. in a human or animal subject or in a biological sample under study or treatment. The effector units or parts of them may remain linked to the target 25 ing units, or they may be partly or completely hydrolyzed or otherwise disinte grated from the latter, either by a spontaneous chemical reaction or equilibrium or by a spontaneous enzymatic process or other biological process, or as a re sult of an intentional operation or procedure such as the administration of hy drolytic enzymes or other chemical substances. It is also possible that the en 30 zymatic process or other reaction is caused or enhanced by the administration of a targeted substance such as an enzyme in accordance with the present in vention. One possibility is that the effector units or parts thereof are hydro lyzed from the targeting agent and/or hydrolyzed into smaller units by the ef 35 fect of one or more of the various hydrolytic enzymes present in tumors (e.g., WO 2004/031218 PCT/F12003/000723 25 intracellularly, in the cell membrane or in the extracellular matrix) or in their near vicinity. Taking into account that the targeting according to the present in vention may be very rapid, even non-specific hydrolysis that occurs every 5 where in the body may be acceptable and usable for hydrolysing one or more effector unit(s) intentionally, since such hydrolysis may in suitable cases (e.g., steric hindrance, or even without any such hindering effects) be so slow that the targeting agents are safely targeted in spite of the presence of hydrolytic enzymes of the body, as those skilled in the art very well understand. The for 10 mation of insoluble products and/or products rapidly absorbed into cells and/or bound to their surfaces after hydrolysis may also be beneficial for the targeted effector units and/or their fragments etc. to remain in the tumors or their closest vicinity. In one preferred embodiment of the invention, the effector units may 15 comprise structures, features, fragments, molecules or the like that make pos sible, cause directly or indirectly, an "amplification" of the therapeutic or other effect, of signal detection, of the binding of preselected substances, including biological material, molecules, ions, microbes or cells. Such "amplification" may, for example, be based on one or more of 20 the following non-limiting types: - the binding, by the effector units, of other materials that can further bind other substances (for example, antibodies, fluorescent antibodies, other "la belled" substances, substances such as avidin, preferably so that several molecules or "units" of the further materials can be bound per each effector 25 unit; - the effector units comprise more than one entity capable of binding e.g. a protein, thus making direct amplification possible; - amplification in more than one steps. Preferred effector units according to the present invention may be 30 selected from the follwing group: cytostatic or cytotoxic agents apoptosis causing or enhancing agents _ enzymes or enzyme inhibitors _ antimetabolites 35 - agents capable of disturbing membrane functions - radioactive or paramagnetic substances WO 2004/031218 PCT/F12003/000723 26 substances comprising one or more metal ions - substances comprising boron, gadolinium, litium - substances suitable for neutron capture therapy - labelled substances 5 - intercalators and substances comprising them - oxidants or reducing agents - nucleotides and their analogues - metal chelates or chelating agents. In a highly preferred embodiment of the invention, the effector unit 10 comprises alpha emittors. In further preferred embodiments of the invention, the effector units may comprise copper chelates such as trans-bis(salicylaldoximaro) copper(II) and its analogues, or platinum compounds such cisplatin, carboplatin. Different types of structures, substances and groups ar known that 15 can be used to cause or enhance e.g., internalization into cells, including for example RQIKIWFQNRRMKWKK; Penetratin (Prochiantz, 1996), as well as stearyl derivatives (Promega Notes Magazine, 2000). As an apoptosis-inducing structure, for example, the peptide se quence KLAKLAK that interacts with mitochondrial membranes inside cells, 20 can be included Ellerby et al. (1999). For use in embodiments of the present invention that include cell sorting and any related applications, the targeting units and agents of the in vention can, for example, be used a) coupled or connected to magnetic particles, 25 b) adsorbed, coupled, linked or connected to plastic, glass or other solid, po rous, fibrous material-type or other surface(s) and the like, c) adsorbed, covalently bonded or otherwise linked, coupled or connected into or onto one or more substance(s) or material(s) that can be used in columns and related systems 30 d) adsorbed, covalently bonded or otherwise linked, coupled or connected into or onto one or more substance(s) or material(s) that can be precipitated, centri fuged or otherwise separated or removed. OPTIONAL UNITS OF THE TARGETING AGENTS ACCORDING TO THE PRESENT INVENTION 35 The targeting agents and targeting units of the present invention may optionally comprise further units, such as: WO 2004/031218 PCT/F12003/000723 27 linker units coupling targeting units, effector units or other optional units of the present invention to each other; solubility modifying units for modifying the solubility of the targeting agents or their hydrolysis product; 5 stabilizer units stabilizing the structure of the targeting units or agents during synthesis, modification, processing, storage or use in vivo or in vitro; charge modifying units modifying the electrical charges of the targeting units or agents or their starting materials; spacer units for increasing the distance between specific units of the targeting 10 agents or their starting materials, to release or decrease steric hindrance or structural strain of the products; reactivity modifyer units; internalizing units or enhancer units for enhancing targeting and uptage of the targeting agents; 15 adsorption enhancer units, such as fat or water soluble structures enhancing absorption of the targeting agents in vivo; or other related units. A large number of suitable linker units are known in the art. Exam ples of suitable linkers are: 20 1. for linking units comprising amino groups: cyclic anhydrides, dicarboxylic or multivalent, optinally activated or derivatized, carboxylic acids, compounds with two or more reactive halogens or compounds with at least one reactive halogen atom and at least one carboxyl group; 2. for linking units comprising carboxyl groups or derivatives thereof: com 25 pounds with at least two similar or different groups such as amino, substi tuted amino, hydroxyl, -NHNH 2 or substituted forms thereof, other known groups for the purpose (activators may be used); 3. for linking an amino group and a carboxyl group: for example amino acids and their activated or protected forms or derivatives; 30 4. for linking a formyl group or a keto group to another group are: a compound comprising e.g. at least one -N-NH 2 or -0-NH 2 or =N-NH 2 or their like; 5. for linking several amino-comprising units: polycarboxylic substances such as EDTA, DTPA and polycarboxylic acids, anhydrides, esters and acyl halides; 35 6. for linking a substance comprising an amino group to a substance comprising either a formyl group or a carboxyl group: hydrazinocarboxylic WO 2004/031218 PCT/F12003/000723 28 acids or their like, preferably so that the hydrazino moiety or the carboxyl group is protected or activated, such as 4-(FMOC-hydrazino)benzoic acid; 7. for linking an organic structure to a metal ion: substances that can be coupled to the organic structure (e.g. by virtue of their COOH groups or 5 their NH 2 groups) or that are integral parts of it, and that in addition comprise a polycarboxylic part for example an EDTA- or DTPA-like structure, peptides comprising several histidines or their like, peptides comprising several cysteines or other moieties comprising an -SH group each, and other chelating agents that comprise functional groups that can 10 be used to link them to the organic structure. A large variety of the above substances and other types of suitable linking agents are known in the art. A large number of suitable solubility modifier units are known in the art. Suitable solubility modifier units comprise, for example: 15 - for increasing aqeous solubility: molecules comprising S0 3 , O-SO3, COOH, COO-, NH 2 , NH 3 ', OH groups, guanidino or amidino groups or other ionic and ionizable groups and sugar-type structures; - for increasing fat solubility or solubility in organic solvents: units comprising (long) aliphatic branched or non-branched alkyl and alkenyl groups, cyclic non 20 aromatic groups such as the cyclohexyl group, aromatic rings and steroidal structures. A large number of units known in the art can be used as stabilizer units, e.g. bulky structures (such as tert-butyl groups, naphthyl and adamantyl and related radicals etc.) for increasing steric hindrance, and D-amino acids 25 and other unnatural amino acids (including p-amino acids, r>-amino acids, amino acids with very large side chains etc.) for preventing or hindering enzy matic hydrolysis. Units comprising positive, negative or both types of charges can be used as charge modifier units, as can also structures that are converted or can 3o be converted into units with positive, negative or both types of charges. Spacer units may be very important, and the need to use such units depends on the other components of the structure (e.g. the type of biologically active agents used, and their mechanisms of action) and the synthetic proce dures used 35 Suitable spacer units may include for example long aliphatic chains or sugar-type structures (to avoid too high lipophilicity), or large rings. Suitable WO 2004/031218 PCT/F12003/000723 29 compounds are available in the art. One preferred group of spacer units are o amino acids with long chains. Such compounds can also be used (simultane ously) as linker units between an amino-comprising unit and a carboxyl comprising unit. Many such compounds are commercially available, both as 5 such and in the forms of various protected derivatives. Units that are susceptible to hydrolysis (either spontaneous chemi cal hydrolysis or enzymatic hydrolysis by the body's own enzymes or en zymes administered to the patient) may be very advantageous in cases where it is desired that the effector units are liberated from the targeting agents e.g. 10 for internalization, intra- or extracellularl DNA or receptor binding. Suitable units for this purpose include, for example, structures comprising one or more ester or acetal functionality, Various proteases may be used for the purposes mentioned. Many groups used for making pro-drugs may be suitable for the purpose of increasing or causing hydrolysis, lytic reactions or other decomposi 15 tion processes. The effector units, the targeting units and the optional units accord ing to the present invention may simultaneously serve more than one function. Thus, for example, a targeting unit may simultaneously be an effector unit or comprise several effector units; a spacer unit may simultaneously be a linker 20 unit or a charge modifier unit or both; a stabilizer unit may be an effector unit with properties different from those of another effector unit, and so on.An effec tor unit may, for example, have several similar or even completely different functions. In one preferred embodiment of the invention, the tumor targeting 25 agents comprise more than one different effector units. In that case, the effec tor units may be, for example, diagnostic and therapeutic units. Thus, for ex ample, it is preferred to use, for boron neutron capture therapy, such agents whose effector units, in addition to comprising boron atoms, also can be de tected or quantified in the patient in vivo after administration of the agent, in so order to be able to ascertain that the agent has accumulated adequately in the tumor to be treated, or to optimize the timing of the neutron treatment, and so on. This goal may be achieved e.g. by using such a targeting agents according to the invention that comprise an effector unit comprising boron atoms (pref erably isotope-enriched boron) and groups detectable e.g. by NMRI. Likewise, 35 the presence of more than one type of therapeutically useful effector units may also be preferred, In addition, the targeting units and targeting agents may, if WO 2004/031218 PCT/F12003/000723 30 desired, be used in combination with one or more "classical" or other tumor therapeutic modalities such as surgery, chemotherapy, other targeting modali ties, radiotherapy, immunotherapy etc. PREPARATION OF TARGETING UNITS AND AGENTS ACCORDING TO 5 THE PRESENT INVENTION The targeting units according to the present invention are preferably synthetic peptides. Peptides can be synthesized by a large variety of well known techniques, such as solid-phase methods (FMOC-, BOC-, and other protection schemes, various resin types), solution methods (FMOC, BOC and 10 other variants) and combinations of these. Even automated appara tuses/devices for the purpose are available commercially, as are also routine synthesis and purification services. All of these approaches are very well known to those skilled in the art. Some methods and materials are described, for example, in the following references: 15 Bachem AG, SASRIN TM (1999), The BACHEM Practise of SPPS (2000), Bachem 2001 catalogue (2001), Novabiochem 2000 Catalog (2000), Peptide and Peptidomimetic Synthesis (2000) and The Combinatorial Chemis try Catalog & Solid Phase Organic Chemistry (SPOC ) Handbook 98/99. Pep tide synthesis is exemplified also in the Examples. 20 As known in the art, it is often advisable, important and/or necessary to use one or more protecting groups, a large variety of which are known in the art, such as FMOC, BOC, and trityl groups and other protecting groups mentioned in the Examples. Protecting groups are often used for protecting amino, carboxyl, hydroxyl, guanyl and -SH groups, and for any 25 reactive groups/functions. As those skilled in the art well know, activation often involves carboxyl function activation and/or activation of amino groups. Protection may also be orthogonal and/or semi/quasi/pseudo orthogonal. Protecting and activating groups, substances and their uses are 30 exemplified in the Examples and are described in the references cited herein, and are also described in a large number of books and other sources of information commonly known in the art (e.g. Protective Groups in Organic Synthesis, 1999). Resins for solid-phase synthesis are also well known in the art, and 35 are described in the Examples and in the above-cited references.

WO 2004/031218 PCT/F12003/000723 31 Cyclic structures according to the present invention may be synthe sized, for example, by methods based on the use of orthogonally protected amino acids. Thus, for example, one amino acid containing an orthogonally protected "extra" COOH function (e.g the (-allyl ester of 5 N-(-FMOC-L-glutamic-acid, i.e., "FMOC-Glu-Oall"), or the (-tert-butyl ester of N-(-FMOC-L-glutamic acid ("FMOC-Glu-OtBu), or the (-4{N-[1 -(4,4-dimethyl-2,6-dioxocyclohexylidene)-3-methylbutyl]-amino}benzyl ester of N-(-FMOC-L-glutamic acid ("FMOC-Glu-Odmab") or the (-2-phenylisopropyl ester of N-(-FMOC-L-glutamic acid 10 ("FMOC-Glu(O-2-PhiPr)-OH"), or related derivatives of other dicarboxylic amino acids, such as aspartic acid; or resin-bound forms of any of the afore mentioned), and one amino acid with an orthogonally protected "extra" amino group (e.g N-(-FMOC-N-(-4-methyltrityl-L-lysine ("FMOC-Lys(Mtt)-OH") or the corresponding derivative of ornithine or some other diaminocarboxylic acid or a 15 resin-bound form of one of these; resin-bound forms, however, not simultane ously with resin-bound forms of the orthogonally protected amino acids with "extra" COOH), may be incorporated in the structure and, after deprotection, the carboxyl and amino groups may be reacted, usually using activator(s). This type of methodology is well known and is described, for example, in the follow 20 ing references Novabiochem Catalog (2000), pp. 19-21 and 33 and specifically B9-B15, and in the references therein, Bachem 2001 catalogue (2001), pp. 31-32, Chan et al. (1995), Yue et al. (1993) and Hirschmann et al. (1998). Suitable starting materials are available commercially, and further ones can be made by methods known in the art. D-amino acid derivatives can 25 also be used in this methodology. Instead of "truly" orthogonal protective groups, also quasiorthogonal/semi-orthogonal/pseudoorthogona protecting groups can be employed, as those skilled in the art understand. Cyclic products made according to the above described methods are usually especially stable in biological milieu, and are thus preferred. This 30 type of structures may be produced by any of the methods for the production of such structures (chemical, enzymatic or biological). Many such methods are well known for those skilled in the art. Cyclic structures of this type can be syntesized chemically with the aid of solid-phase synthesis but they can like wise be synthesized using solution methods or a combination of both, as those 35 skilled in the art well know. Amino acids with an "extra" carboxyl or amino func- WO 2004/031218 PCT/F12003/000723 32 tion suitable for cyclization purposes (when adequately protected) include (as non-limiting possibilities), for example, those with the structures shown below: COOH COOH COOH COOH

NH

2

CH-NH

2

CH-NH

2

H

2 N NH2

(CH

2 )n (CH 2 )n COOH CH 2

-NH

2 COOH 5 In solution cyclizations of any type, dilute solutions are normally ad vantageous, as is well known by those skilled in the art. The targeting units and agents according to the present invention may also be prepared as fusion proteins or by other suitable recombinant DNA 1o methods known in the art. Such an approach for preparing the peptides ac cording to the present invention is preferred especially when the effector units and/or other optional units are peptides or proteins. One example of a useful protein effector unit is glutathion-S-transf erase (GTS). ADVANTAGES OF THE TARGETING UNITS AND TARGETING AGENTS OF 15 THIS INVENTION There are acknowledged problems related to peptides intended for diagnostic or therapeutic use. One of these problems stem from the length of the sequence: the longer it grows, the more difficult or even impossible the synthesis of the desired product becomes, especially if there are other 20 synthetic problems such as the presence of difficult residues that require protection-deprotection and/or cause side reactions etc. The tendency to side rections, and possible synthesis termination (that not only decreases the yield of the desired product if this is formed at all, but also gives rise to products with a wrong length of the peptide chain) and formation of serious amounts of 25 harmful by-products is drastically increased by the presence in the desired sequence of any amino acid(s) that require(s) side-chain protection (e.g., basic side-chains such as those of lysine, histidine and tryptophan) and (of course) also deprotection. All of these problems also make, as those skilled in the art very well know, the purification of the desired peptides very much more difficult 30 and may make production of adequately purified material impossible.

WO 2004/031218 PCT/F12003/000723 33 As compared to known products that contain long and difficult-to make sequences with problematic amino acid residues, the peptides of the present invention are clearly superior, as described in more detail below. Thus, the products and methods of the present invention and their 5 use offer highly significant and very important advantages over the prior art. The targeting units of this invention can be synthesized easily and reliably. An advantage as compared to many prior art peptides is that the targeting units and motifs of this invention do not need to comprise the problematic basic amino acids lysine and histidine, nor tryptophan, all of which 10 may cause serious side-reactions in peptide synthesis, and, due to which the yield of the desired product might be lowered radically or even be impossible to obtain in adequate amounts or with adequate quality. When present, histidine, lysine and tryptophan must be adequately protected using suitable protecting groups that remain intact during the 15 synthesis prodecures. This may be very difficult and at least increases the costs and technical problems. Also costs are remarkably increased by the reagents and work-load and other costs of the deprotection steps and the costs per unit of desired product may be increased. Because of their smaller size and thus drastically less steps in the 20 synthesis, the peptides of the present invention are much easier and cheaper to produce than targeting peptides of the prior art. As histidine is not needed in the products of the present invention, the risk of racemization of it is of no concern. It is a great advantage not only for the economic synthesis of the 25 products of the present invention but also for the purification and analysis and quality control that any racemization of histidine is outside consideration. It also makes any administration to humans and animals safer and more straightforward. Because of their smaller size, the peptides of the present invention 30 can also be purified much more reliably and easily and with much less labor and apparatus-time, and thus with clearly lower costs. Overall costs are thus drastically reduced and better products can be obtained and in greater amounts. Furthermore, the reliability of the purification is much better, giving less concern of toxic remainders and of fatal or otherwise serious side-effects 35 in therapeutic and diagnostic applications.

WO 2004/031218 PCT/F12003/000723 34 Shorter synthesis protocols with relatively few steps produce less impurities, making the peptides of the present invention highly advantageous. The risks of toxic and even fatal impurities, allergens etc. are dramatically lowered and, in addition, purification is easier. 5 The analysis and thus the quality control of the products of the present invention is easier and less costly, than that of the longer and more 'difficult' peptide sequences. This increases the reliability of the analyses and of quality control. As residues such as lysine are not present in the targeting unit, 10 there is no the risk of the effector units being inadequately connected to such residues. This is a remarkable advantage. The effector unit can easily be linked to the peptides and peptidyl analogues and peptidomimetic substances of the present invention using (outside the targeting motif) for example protected lysine or ornithine as there 15 is no risk of simultaneous reaction of any lysine residue in the targeting motif. For cyclization of the peptides of the present invention, protected lysine or ornithine can be used, as the targeting units do not contain such amino acids. This is an enormous advantage The lactam bridge is clearly superior to the commonly used 20 disulphide bridge. This is due to a number of reasons, such as: 1. Lactams are highly stable against reducing agents such as thiols that easily destroy disulphide bridges and can cause various undesired reactions, for example dimerization, polymerization and formation of disulphide bridges to other thiols; 25 2. Lactams can be synthesized in a highly controlled fashion for example by using orthogonal protection; 3. Lactams are more stable than disulphides in biological milieu. 4. The purification of lactams is easier than with disulphides. 5. Because of lack of reaction with macromolecules having SH 30 groups lactones probable are less prone to cause allergic reactions. In solid synthesis of targeting agents according to the present invention, the effector units and optional additional units may be linked to the targeting peptide when still connected to the resin without the risk that the removal of the protecting groups will cause destruction of additional unit. 35 Similar advantage applies to solution syntheses.

WO 2004/031218 PCT/F12003/000723 35 Another important advantage of the present invention and its prod ucts, methods and uses according to it is the highly selective and potent tar geting of the products. As compared to targeted therapy using antibodies or antibody frag 5 ments, the products and methods of in the present invention are highly advan tageous because of several reasons. Potential immunological and related risks are also obvious in the case of large biomolecules. Allergic reactions are of great concern with such products, in contrats to small synthetic molecules such as the targeting agents, units and motifs of the present invention. 10 As compared to targeting antibodies or antibody fragments, the products and methods described in the present invention are highly advanta geous because their structure can be easily modified if needed or desired. Specific amino acids such as histidine, tryptophan, tyrosine, threonine can be omitted if dersired, and very few functional groups are necessary. On the other 15 hand, it is possible, without disturbing the targeting effect, to include various different structural units, to specific desired properties that are of special value in specific applications. USE OF TARGETING AGENTS ACCORDING TO THE PRESENT INVEN TION 20 The targeting units and targeting agents according to the present invention are useful in cancer diagnostics and therapy, as they selectively tar get to tumors in vivo, as shown in the examples. The effector unit may be cho sen according to the desired effect, detection or therapy. The desired effect may also be achieved by including the effector in the targeting unit as such. 25 For use in radiotherapy the targeting unit itself may be e.g., radioactively la belled. The present invention also relates to diagnostic compositions com prising an effective amount of at least one targeting agent according to the present invention. In addition to the targeting agent, a diagnostic composition 30 according to the present invention may, optionally, comprise carriers, solvents, vehicles, suspending agents, labelling agents and other additives commonly used in diagnostic compositions. Such diagnostic compositions are useful in diagnosing tumors, tumor cells and metastasis. A diagnostic composition according to the present invention may be 35 formulated as a liquid, gel or solid formulation, preferably as an aqueous liquid, containing a targeting agent according to the present invention in a concentra- WO 2004/031218 PCT/F12003/000723 36 tion ranging from about 0.00001 pig/l to 25 x 10 7 ptg/l. The compositions may further comprise stabilizing agents, detergents, such as polysorbates and Tween, as well as other additives. The concentrations of these components may vary significantly depending on the formulation used. The diagnostic 5 compositions may be used in vivo or in vitro. The present invention also includes the use of the targeting agents and targeting units for the manufacture of pharmaceutical compositions for the treatment of cancer. The present invention also relates to pharmaceutical compositions io comprising a therapeutically effective amount of at least one targeting agent according to the present invention. The pharmaceutical compositions may be used to treat, prevent or ameliorate cancerdiseases, by administering an therapeutically effective dose of the pharmaceutical composisiton comprising targeting agents or targeting units according to the present invention or thera 15 peutically acceptable salts, esters or other derivatives thereof. The composi tions may also include different combinations of targeting agents and targeting units together with labelling agents, imaging agents, drugs and other additives. A therapeutically effective amount of a targeting agent according to the present invention may vary depending on the formulation of the pharma 20 ceuticakl composition. Preferably, a composition according to the present in vention may comprise a targeting agent in a concentration varying from about 0.00001 tg/l to 250 g/l, more preferably about 0,001 pg/l to 50 g/l, most pref erably 0,01 pg/l to 20 g/l. A pharmaceutical composition according to the present invention is 25 useful for administration of a targeting agent according to the present inven tion. Pharmaceutical compositions suitable for peroral use, for intravenous or local injection, or infusion are particularly preferred. The pharmaceutical com positions may be used in vivo or ex vivo. The preparations may be lyophilized and reconstituted before ad 30 ministration or may be stored for example as a solution, solutions, suspen sions, suspension-solutions etc. ready for administration or in any form or shape in general, including powders, concentrates, frozen liquids, and any other types. They may also consist of separate entities to be mixed and, pos sibly, otherwise handled and/or treated etc. before use. Liquid formulations 35 provide the advantage that they can be administered without reconstitution. The pH of the solution product is in the range of about 1 to about 12, prefera- WO 2004/031218 PCT/F12003/000723 37 bly close to physiological pH. The osmolality of the solution can be adjusted to a preferred value using for example sodium chloride and/or sugars, polyols and/or amino acids and/or similar components. The compositions may further comprise pharmaceutically acceptable excipients and/or stabilizers, such as 5 albumin, sugars and various polyols, as well as any acceptable additives, or other active ingredients such as chemotherapeutic agents. The present invention also relates to methods for treating cancer, especially solid tumors by administering to a patient in need of such treatment a therapeutically efficient amount of a pharmaceutical composition according 10 to the present invention. Therapeutic doses may be determined empirically by testing the targeting agents and targeting units in available in vitro or in vivo test systems. Examples of such tests are given in the examples. Suitable therpeutically ef fective dosage may then be estimated from these experiments. 15 For oral administration it is important that the targeting units and targeting agent are stable and adequately absorbed from the intestinal tract. The pharmaceutical compositions according to the present invention may be administered systemically, non-systemically, locally or topically, par enterally as well as non-parenterally, e.g. subcutaneously, intravenously, in 20 tramuscularly, perorally, intranasally, by pulmonary aerosol, by injection or in fusion into a specific organ or region, buccally, intracranically or intraperito neally. Amounts and regimens for the administration of the tumor targeting agents according to the present invention can be determined readily by those 25 with ordinary skill in the clinical art of treating cancer. Generally, the dosage will vary depending upon considerations such as: type of targeting agent em ployed; age; health; medical conditions being treated; kind of concurrent treatment, if any, frequency of treatment and the nature of the effect desired; gender; duration of the symptoms; and, counterindications, if any, and other 30 variables to be adjusted by the individual physician. Preferred doses for ad ministration to human patients targeting targeting units or agents according to the present invention may vary from about 0.000001 .tg to about 40 mg per kg of body weight as a bolus or repeatedly, e.g., as daily doses. The targeting units and targeting agents and pharmaceutical com 35 positions of the present invention may also be used as targeting devices for delivery of DNA or RNA or structural and functional analogues thereof, such as WO 2004/031218 PCT/F12003/000723 38 phosphorothioates, or peptide nucleic acids (PNA) into tumors and their metas tases or to isolated cells and organs in vitro; i.e. as tools for gene therapy both in vivo and in vitro. In such cases the targeting agents or targeting units may be parts of viral capsids or envelopes, of liposomes or other "containers" of 5 DNA/RNA or related substances, or may be directly coupled to the DNA/RNA or other molecules mentioned above. The present invention also includes kits and components for kits for diagnosing, detecting or analysing cancer or cancer cells in vivo and in vitro. Such kits comprise at least a targeting agent or targeting unit of this invention 10 together with diagnostic entities enabling detection. The kit may comprise for example a targeting agent and/or a targeting unit coupled to a unit for detec tion by e.g. immunological methods, radiation or enzymatic methods or other methods known in the art. Further, the targeting units and agents of this invention as well as 15 the targeting motifs and sequences can be used as lead compounds to design peptidomimetics for any of the purposes described above. Yet further, the targeting units and agents as well as the targeting motifs and sequences of the present invention, as such and/or as coupled to other materials, can be used for the isolation, purification and identification of 20 the cells, molecules and related biological targets. The following non-limiting examples illustrate the invention further. EXAMPLES A list of reagents used in the examples below and reagent suppliers is included after the last numbered example. 25 EXAMPLE 1 SYNTHESIS OF TARGETING UNIT (PEPTIDE) LRS The functionally protected, resin bound targeting unit (protected peptide), comprising targeting motif LRS (leucyl arginyl serine), was synthe sized by means of manual synthesis as described in Example 2 below. 30 The following reagents were employed as starting materials (in this order): Fmoc-Ser(tBu) Resin Fmoc-L-Arg(Pbf)-OH Fmoc-L-Leu-OH WO 2004/031218 PCT/F12003/000723 39 After the last cycle of the coupling process, a small sample of the resin (comprising the still fully protected peptide) was subjected to the treat ment described in Example 2 for Fmoc removal (steps 1-10 in that Example), after which the sample of peptide was cleaved from the resin by three hours' 5 treatment with the cleavage mixture described in Example 2, and isolated as described in the same Example. Then, the product (LRS) was identified with the aid of its positive mode MALDI-TOF mass spectrum, in which the M+1 ion of LRS was clearly predominant. 10 MALDI-TOF data (LRS): calculated molecular mass = 374.44 observed signals: 375.30 M+H 397.22 M+Na 15 EXAMPLE 2 GENERAL PROCEDURES FOR PEPTIDE SYNTHESIS: MANUAL SOLID PHASE SYNTHESES. MASS SPECTRAL MEASUREMENTS. All synthetic procedures were carried out in a sealable glass funnel equipped with a sintered glass filter disc of porosity grade between 2 and 4, a 20 polypropene or phenolic plastic screw cap on top (for sealing), and two PTFE key stopcocks: one beneath the filter disc (for draining) and one at sloping an gle on the shoulder of the screw-capped neck (for argon gas inlet). The funnel was loaded with the appropriate solid phase synthesis resin and solutions for each treatment, shaken powerfully with the aid of a 25 "wrist movement" bottle shaker (Gallenkamp) for an appropriate period of time, followed by filtration effected with a moderate argon gas pressure. The general procedure of one cycle of synthesis (= the addition of one amino acid unit) was as follows: The appropriate Wang resin (Applied Biosystems), loaded with ap 30 proximately 1 mmol of Fmoc-peptide (= peptide whose amino-terminal amino group was protected with the 9-fluorenylmethyloxycarbonyl group) consisting of two or more amino acid units, or with approximately 1 mmol of the appropri ate Fmoc-amino acid (i.e., amino acid carrying the aforementioned protecting group; approximately 2g of resin, 0.5 mmol/g) was treated in the way described WO 2004/031218 PCT/F12003/000723 40 below, each treatment step comprising shaking for 2.5 minutes with 30 ml of the solution or solvent indicated and filtration if not mentioned otherwise. 'DCM' means shaking with dichloromethane, and 'DMF' means shaking with NN-dimethylformamide (DMF may be replaced by NMP, i.e. N 5 methylpyrrolidinone). The steps of the treatment were: 1. DCM, shaking for 10-20 min 2. DMF 3. 20% (by volume) piperidine in DMF for 5 min 10 4. 20% (by volume) piperidine in DMF for 10 min 5. to 7. DMF 8. to 10. DCM 11. DMF 12. DMF solution of 3 mmol of activated amino acid (preparation de 15 scribed below), shaking for 2 hours 13. to 15. DMF 16. to 18. DCM After the last treatment (18) argon gas was led through the resin for approximately 15 min and the resin was stored under argon (in the sealed re 20 action funnel if the synthesis was to continue with further units). Activation of the 9-fluorenylmethyloxycarbonyl-N-protected amino acid (Fmoc-amino acid) to be added to the amino acid or peptide chain on the resin was carried out, using the reagents listed below, in a separate vessel prior to treatment step no. 12. Thus, the Fmoc-amino acid (3 mmol) was dis 25 solved in approximately 10 ml of DMF, treated for 1 min with a solution of 3 mmol of HBTU dissolved in 6 ml of a 0.5 M solution of HOBt in DMF, and then immediately treated with 3 ml of a 2.0 M DIPEA solution for 5 min. The activation reagents used for activation of the Fmoc-amino acid were as follows: 3o HBTU = 2-(1H-benzotriazol-1-yl)-1,1,3,3-tetramethyluronium hexafluoro phosphate, CAS No. [94790-37-1], Applied Biosystems Cat. No. 401091, mo lecular weight: 379.3 g/mol HOBt = 1-Hydroxybenzotriazole, 0.5 M solution in DMF, Applied Biosystems Cat. No. 400934 35 DIPEA = N,N-Diisopropylethylamine, 2.0 M solution in N-methylpyrrolidone, Applied Biosystems Cat. No. 401517 WO 2004/031218 PCT/F12003/000723 41 The procedure described above was repeated in several cycles us ing the appropriate different Fmoc-amino acids, carrying suitable protecting group(s), to produce a resin-bound source of the appropriate peptide (i.e., a "resin-bound" peptide). The procedure provides also a practical way of con 5 necting certain effector and/or spacer and/or linker units and so on, for in stance biotin or the Fmoc-Ahx (= 6-(Fmoc-amino)-hexanoy) moiety, to the re sin-bound peptide. Cleavage from the resin was carried out using the following reagent mixture: 10 trifluoroacetic acid (TFA) 92.5 vol-% water 5.0 vol-% ethanedithiol 2.5 vol-%. After the removal of the protecting Fmoc group via steps 1. to 10. (as described in the general procedure above), the resin was treated with three 15 portions of the above reagent mixture (each about 15 ml for 1 g of the resin), each for one hour. The treatments were carried out under argon atmosphere in the way described above. The TFA solutions obtained by filtration were then concentrated under reduced pressure using a rotary evaporator and were re charged with argon. Some diethyl ether was added and the concentration re 20 peated. The concentrated residue was allowed to precipitate overnight under argon in dietyl ether in a refrigerator. The supernatant ether was removed and the precipitate rinsed with diethyl ether. For mass spectrum (MALDI-TOF+) de termination, a sample of the precipitate was dissolved in solvents adequate for the spectral method, followed by filtration and, as needeed, dilution of the fil 25 tered solution. Further purification was done using reversed phase high performance liquid chromatographic (HPLC) methods by means of a "Waters 600" pump apparatus using a C-18 type column of particle size 10 microme ters and a linear eluent gradient whose composition was changed during 30 minutes from 99.9% water/0.1% TFA to 99.9% acetonitrile/0.1% TFA. The di so mensions of the HPLC columns were 25 cm x 21.2 mm (Supelco cat. no. 567212-U) and 15 cm x 10 mm (Supelco cat. no. 567208U). Detection was based on absorbance at 218 nm and was carried out using a "Waters 2487" instrument. The cleavage mixture described above also simultaneously re 35 moved the following protecting groups: trityl (Trt) as used for cysteine -SH pro tection; 2,2,4,6,7-pentamethyldihydrobenzofu ran-5-sulfonyl (Pbf) as used for WO 2004/031218 PCT/F12003/000723 42 protection of side chain of arginine; the tert-butyl group (as an ester group on the carboxyl function; OtBu) as used for protection of the side-chain carboxyl group of glutamic acid and/or aspartic acid, and can normally be used also for removal of these protecting groups on analogous structures (thiol, guanyl, car 5 boxyl). It did not cause Fmoc removal. The cleavage procedure described above can be carried out also without the removal of the Fmoc group, to produce the amino terminal N Fmoc-derivative of the peptide, or for a peptide linked to an effector unit (com prising no Fmoc). 10 MASS SPECTRAL METHOD EMPLOYED Matrix Assisted Laser Desorption Ionization - Time of Flight (MALDI -TOF) Type of the intrument: Bruker Biflex MALDI TOF mass spectrometer Supplier of the instrument: 15 Bruker Daltonik GmbH Fahrenheitstrasse 4 D-28359 Bremen Germany MALDI-TOF POSITIVE ION REFLECTOR MODE 20 External standards: Angiotensin 11 and ACTH(18-39) Matrix: alpha-cyano-4-hydroxycinnamic acid (saturated solution in aqueous 50% ace tonitrile containing 0.1% of trifluoroacetic acid). 25 The sample, together with the matrix, was dried onto the target plate under a gentle stream of warm air. MALDI-TOF NEGATIVE ION REFLECTOR MODE: External standards: cholecystokinin and glucagon 30 Matrix: 2,4,6-trihydroxyacetophenone (3 mg/ml in 10 mM ammonium citrate in 50% acetonitrile). The sample, mixed with the matrix, was immediately dried onto the target plate under vacuum.

WO 2004/031218 PCT/F12003/000723 43 SAMPLE PREPARATION The specimen was mixed at a 10-100 picomol/microliter concentra tion with the matrix solution as described. "Shooting" by nitrogen laser at wawelength 337 nm. The voltage of 5 the probe plate was 19 kV in the positive ion reflector mode and -19 kV in the negative ion reflector mode. GENERAL REMARKS ABOUT THE SPECTRA (CONCERNING POSITIVE ION MODE ONLY) In all cases the M+1 (i.e. the one proton adduct M+H+) signal with 10 its typical fine structure based on isotope satellites was clearly predominant. In almost all cases, the M+1 signal pattern was accompanied by a similar but markedly weaker band of peaks at M+23 (Na+ adduct). In addition to the bands at M+1 and M+23, also bands at M+39 (K+ adduct) or M+56 (Fe+ ad duct) could be observed in many cases. 15 In case of substances with a low molecular mass, the 'matrix sig nals' (signals due to the constituents of the matrix/ 'the ionization environment') have been omitted (i.e., signals at 294 and 380 Da have been omitted). The calculated molecular mass values reported within synthesis ex amples correspond to the most abundant isotopes of each element, i.e., the 20 'exact masses'. The interpretations given for signals are only tentative. EXAMPLE 3 GENERAL PROCEDURES FOR 12-PROMOTED CYCLIZATION OF PEPTI DE/TARGETING UNIT OR TARGETING AGENT ON RESIN (FOR PEPTIDES AND TARGETING UNITS AND TARGETING AGENTS COMPRISING CYS 25 TEINES) The resin (1 g) was swelled on CH2Cl2 (15 ml) and stirred for 20 minutes. The solvent was removed by filtration and the resin was treated once with DMF (15 ml) for three minutes. After filtration, the resin-bound peptide (or targeting agent) was treated with iodine (5 molar equivalents) in DMF (10 ml) 30 for 1 hour. The DMF-iodine solution was removed by filtration and the residue was washed three times with DMF (15 ml) and three times with CH2CI2 (15ml) for 3 minutes each time.

WO 2004/031218 PCT/F12003/000723 44 In case that a 'plain' peptide (without the Fmoc group) was to be prepared, the Fmoc group was removed and the peptide was released from the resin according to the general procedure described in Example 2 and puri fied by reversed phase HPLC. In the case of targeting agents comprising no 5 Fmoc group, the product was released from the resin and purified analogously. Material used: Iodine CAS No.7553-56-2 molecular weight: 253.81 10 Merck Art. No. 4760 EXAMPLE 4 SYNTHESIS OF TARGETING UNIT (PEPTIDE) DLRSK The functionally protected, resin bound targeting unit (protected peptide), comprising targeting motif LRS, was synthesized by means of man 15 ual synthesis as described in Example 2 above. The following reagents were employed as starting materials (in this order): Fmoc-Lys(Mtt)-resin Fmoc-L-Ser(tBu)-OH 20 Fmoc-L-Arg(Pbf)-OH Fmoc-L-Leu-OH Fmoc-Asp(2-phenylisopropyl ester)-OH After the last cycle of the coupling process,the still resin-bound tar geting unit was subjected to the cyclization process in which an extra amide 25 bond is formed as described in Example 17. After the cyclization process (macrolactam formation) a small sample of the resin (containing the still fully protected cyclized peptide) was subjected to the treatment described in Exam ple 2 for Fmoc removal (steps 1-10 in that Example), after which the sample of peptide was cleaved from the resin by three hours' treatment with the cleavage 3o mixture described in Example 2, and isolated as described in the same Exam ple. Then, the product (DLRSK macrolactam) was identified with the aid of its positive mode MALDI-TOF mass spectrum, in which the M+1 ion of cyclic DLRSK was clearly predominant. 35 WO 2004/031218 PCT/F12003/000723 45 MALDI-TOF data (cyclic DLRSK): calculated molecular mass = 599.34 observed signals: 600.42 M+H 5 622.40 M+Na 638.29 M+K Fmoc-DLRSK macrolactam Cyclic Fmoc-DLRSK was prepared and identified in analogous manner to cyclic DLRSK with the exeption of the final Fmoc removal that was 10 omitted in this case. MALDI-TOF data (cyclic Fmoc-DLRSK): calculated molecular mass = 821.41 observed signals: 822.60 M+H 15 844.62 M+Na EXAMPLE 5 SYNTHESIS OF TARGETING UNIT (PEPTIDE) DGRGLRSK (CYCLIC BY VIRTUE OF LACTAM BRIDGE) The functionally protected, resin bound targeting unit (protected 20 peptide), comprising targeting motif LRS, was synthesized by means of man ual synthesis as described in Example 2 above. The following reagents were employed as starting materials (in this order): Fmoc-Lys(Mtt) Resin 25 Fmoc-L-Ser(tBu)-OH Fmoc-L-Arg(Pbf)-OH Fmoc-L-Leu-OH Fmoc-Gly-OH Fmoc-L-Arg(Pbf)-OH 3o Fmoc-Gly-OH Fmoc-Asp(2-phenylisopropyl ester)-OH After the last cycle of the coupling process, the still resin-bound tar geting unit was subjected to the cyclization process in which an extra amide bond is formed as described in Example 17. After the cyclization process 35 (macrolactam formation) a small sample of the resin (containing the still fully WO 2004/031218 PCT/F12003/000723 46 protected cyclized peptide) was subjected to the treatment described in Exam ple 2 for Fmoc removal (steps 1-10 in that Example), after which the sample of peptide was cleaved from the resin by three hours' treatment with the cleavage mixture described in Example 2, and isolated as described in the same Exam 5 pie. Then, the product (DGRGLRSK macrolactam) was identified with the aid of its positive mode MALDI-TOF mass spectrum, in which the M+1 ion of cyclic DGRGLRSK was clearly predominant. MALDI-TOF data (cyclic DGRGLRSK): 10 calculated molecular mass = 869.48 observed signal: 870.48 M+H EXAMPLE 6 SYNTHESIS OF TARGETING UNIT (PEPTIDE) DRGLRSK (CYCLIC BY VIR 15 TUE OF LACTAM BRIDGE) The functionally protected, resin bound targeting unit (protected peptide), comprising targeting motif LRS, was synthesized by means of man ual synthesis as described in Example 2 above. The following reagents were employed as starting materials (in this 20 order): Fmoc-Lys(Mtt) Resin Fmoc-L-Ser(tBu)-OH Fmoc-L-Arg(Pbf)-OH Fmoc-L-Leu-OH 25 Fmoc-Gly-OH Fmoc-L-Arg(Pbf)-OH Fmoc-Asp(2-phenylisopropyl ester)-OH After the last cycle of the coupling process, the still resin-bound tar geting unit was subjected to the cyclization process in which an extra amide 30 bond is formed as described in Example 17. After the cyclization process (macrolactam formation) a small sample of the resin (containing the still fully protected cyclized peptide) was subjected to the treatment described in Exam ple 2 for Fmoc removal (steps 1-10 in that Example), after which the sample of peptide was cleaved from the resin by three hours' treatment with the cleavage WO 2004/031218 PCT/F12003/000723 47 mixture described in Example 2, and isolated as described in the same Exam ple. Then, the product (DRGLRSK macrolactam) was identified with the aid of its positive mode MALDI-TOF mass spectrum, in which the M+1 ion of 5 cyclic DRGLRSK was clearly predominant. MALDI-TOF data (cyclic DRGLRSK): calculated molecular mass = 812.46 observed signal: 813.34 M+H 10 EXAMPLE 7 SYNTHESIS OF TARGETING UNIT (PEPTIDE) DRYYNLRSK (CYCLIC BY VIRTUE OF LACTAM BRIDGE) The functionally protected, resin bound targeting unit (protected peptide), comprising targeting motif LRS, was synthesized by means of man 15 ual synthesis as described in Example 2 above. The following reagents were employed as starting materials (in this order): Fmoc-Lys(Mtt) Resin Fmoc-L-Ser(tBu)-OH Fmoc-L-Arg(Pbf)-OH 20 Fmoc-L-Leu-OH Fmoc-L-Asn-OH Fmoc-L-Tyr(tBu)-OH in two subsequent cycles Fmoc-L-Arg(Pbf)-OH again Fmoc-Asp(2-phenylisopropyl ester)-OH 25 After the last cycle of the coupling process, the still resin-bound tar geting unit was subjected to the cyclization process in which an extra amide bond is formed as described in Example 17. After the cyclization process (macrolactam formation) a small sample of the resin (containing the still fully protected cyclized peptide) was subjected to the treatment described in Exam 30 ple 2 for Fmoc removal (steps 1-10 in that Example), after which the sample of peptide was cleaved from the resin by three hours' treatment with the cleavage mixture described in Example 2, and isolated as described in the same Exam ple.

WO 2004/031218 PCT/F12003/000723 48 Then, the product (DRYYNLRSK macrolactam) was identified with the aid of its positive mode MALDI-TOF mass spectrum, in which the M+1 ion of cyclic DRYYNLRSK was clearly predominant. Mass spectral data (cyclic DRYYNLRSK): 5 calculated molecular mass = 1195.61 observed signal: 1096.28 M+H EXAMPLE 8 SYNTHESIS OF TARGETING UNIT (PEPTIDE) DSRYNLRSK (CYCLIC BY 1o VIRTUE OF LACTAM BRIDGE) The functionally protected, resin bound. targeting unit (protected peptide), comprising targeting motif LRS, was synthesized by means of man ual synthesis as described in Example 2 above. The following reagents were employed as starting materials (in this order): 15 Fmoc-Lys(Mtt) Resin Fmoc-L-Ser(tBu)-OH Fmoc-L-Arg(Pbf)-OH Fmoc-L-Leu-OH Fmoc-L-Asn-OH 20 Fmoc-L-Tyr(tBu)-OH Fmoc-L-Arg(Pbf)-OH again Fmoc-L-Ser(tBu)-OH again Fmoc-Asp(2-phenylisopropyl ester)-OH After the last cycle of the coupling process, the still resin-bound tar 25 geting unit was subjected to the cyclization process in which an extra amide bond is formed as described in Example 17. After the cyclization process (macrolactam formation) a small sample of the resin (containing the still fully protected cyclized peptide) was subjected to the treatment described in Exam ple 2 for Fmoc removal (steps 1-10 in that Example), after which the sample of 30 peptide was cleaved from the resin by three hours' treatment with the cleavage mixture described in Example 2, and isolated as described in the same Exam ple. Then, the product (DSRYNLRSK macrolactam) was identified with the aid of its positive mode MALDI-TOF mass spectrum, in which the M+1 ion 35 of cyclic DSRYNLRSK was clearly predominant.

WO 2004/031218 PCT/F12003/000723 49 Mass spectal data (cyclic DSRYNLRSK): calculated molecular mass = 1119.58 observed signal: 1120.24 M+H 5 EXAMPLE 9 SYNTHESIS OF TARGETING UNIT (PEPTIDE) AHXDLRSK The targeting unit has the structure: HN

>-NH

2 0 N 0 0 H N N NNH H H 0 H N

H

2 N 0 H N H N OH 0 OH 0 10 The functionally protected, resin bound targeting unit (protected peptide), comprising targeting motif LRS, was synthesized by means of man ual synthesis as described in Example 2 above. The following reagents were employed as starting materials (in this 15 order): Fmoc-Lys(Mtt)-resin Fmoc-L-Ser(tBu)-OH Fmoc-L-Arg(Pbf)-OH Fmoc-L-Leu-OH 20 Fmoc-Asp(2-phenylisopropyl ester)-OH Fmoc-6-aminohexanoic acid After the last cycle of the coupling process, the still resin-bound tar geting unit was subjected to the cyclization process in which an extra amide bond is formed as described in Example 17. After the cyclization process 25 (macrolactam formation) a small sample of the resin (containing the still fully protected cyclized peptide) was subjected to three hours' treatment with the cleavage mixture described in Example 2. By that way a sample of peptide WO 2004/031218 PCT/F12003/000723 50 was cleaved from the resin and the protecting groups of side chains of that sample were removed with the exeption of the final Fmoc removal that was omitted in this case. The sample was isolated as described in the same Exam ple. 5 Then, the product (DLRSK macrolactam) was identified with the aid of its positive mode MALDI-TOF mass spectrum, in which the M+1 ion of cyclic DLRSK was clearly predominant. MALDI-TOF data (cyclic Fmoc-AhxDLRSK): calculated molecular mass = 938.50 10 observed signal: 939.50 M+1 EXAMPLE 10 SYNTHESIS OF TARGETING AGENT FMOC2DAP-DLRSK (DAP= DIA MINOPROPIONYL), COMPRISING THE EFFECTOR UNIT DIAMINOPROPI 15 ONIC ACID COUPLED (LINKED DIRECTLY, WITHOUT SPECIFIC LINKER UNITS) VIA ITS CARBOXYL GROUP TO THE N-TERMINAL AMINO GROUP OF THE PEPTIDE DLRSK BY VIRTUE OF AN AMIDE BOND, AND ALSO COMPRISING THE TARGETING UNIT DLRSK, THAT IS CYCLIC BY VIRTUE LACTAM BRIDGE 20 The targeting agent was synthesized using manual synthesis as de scribed in Example 2 above (analogously to the synthesis in Example Example 4 above, including cyclization). Next, the sequence DLRSK was continued with DL-2,3-Bis(Fmoc-amino)propionic acid by means of the general coupling met hods described in Example 2. 25 The preparation of DL-2,3-Bis(Fmoc-amino)propionic acid: DL-2,3-diaminopropionic acid monohydrochloride was dissolved in 15 mL of aqueous 10% Na2CO3 solution. Then 7 mL of dioxane was added and the reaction mixture cooled to +4 0 C. Fmoc-chloride in 20 mL of dioxane was added and the reaction mixture stirred for one hour at +40C. After contin 30 ued stirring at room temperature overnight the reaction mixture was extracted with ethyl acetate that was then evaporated. The residue was triturated with n hexane and washed with small amount of hot ethyl acetate to afford white solid that was dried in vacuo overnight.

WO 2004/031218 PCT/F12003/000723 51 Reagents used: DL-2,3-diaminopropionic acid monohydrochloride Fmoc-chloride; 9-fluorenylmethyl chloroformate 98%; C.A.S. no: 28920-43-6 Acros, cat no.: 170940250 5 MALDI-TOF data (Fmoc2Dap-DLRSK, cyclic): calculated molecular mass = 1129.53 observed signal: 1130.32 M+H EXAMPLE 11 10 GENERAL PROCEDURE EMPLOYED IN THE SYNTHESES OF BIOTI NYLATED COMPOUNDS [TARGETING AGENTS COMPRISING ONE D BIOTIN (VITAMIN H) AS AN EFFECTOR UNIT] The appropriate protected peptide was synthesized on using solid phase synthesis according to the general procedure described in Example 2. 15 The peptide was not deprotected and also not removed from the resin. The re sin-bound peptide was added to the reaction flask. The resin was swelled us ing CH2CI2 (15 ml) and stirred for 20 minutes. The solvent was removed by fil tration and the resin was treated once with DMF for three minutes. The peptide was deprotected using 20% piperidine solution in DMF (20ml) and shaking 20 therewith for 5, and the process was repeated using (now shaking for 10 min utes). The resin was washed three times with DMF (15 ml) and three times with CH2Cl2 (15mi) and once with DMF (15 ml) for three minutes each time. D-biotin (3 molar equivalents) in DMF (10 ml) (heterogenous sus pension) was treated in a separate vessel with a 0.5 M solution of 25 HBTU/HOBT in DMF (3 molar eq.) for one minute. Into the vessel was added a 2 M solution of di-isopropylethylamine in NMP (6 molar eq.). After the addition, the reaction mixture became homogenous. The mixture was added to the re action apparatus and the apparatus was shaken for 2 hours. The reaction mixture was then filtered and the residue was washed so three times with DMF (15 ml) and three times with CH2Cl2 (15ml) for 3 min utes each time. In case that the peptide was to be both biotinylated as described herein and cyclized by an iodine treatment as described in Example 3, the cy clization was performed after the biotinylation procedure. 35 WO 2004/031218 PCT/F12003/000723 52 Material used: D-Biotin (Vitamin H) CAS No. 58-85-5 molecular weight: 244.3 5 Sigma B-4501 99% EXAMPLE 12 SYNTHESIS OF TARGETING AGENT BIO-LRS (BIO = D-BIOTIN = VITAMIN H), COMPRISING THE EFFECTOR UNIT D-BIOTIN COUPLED (LINKED DI 10 RECTLY, WITHOUT SPECIFIC LINKER UNITS) VIA ITS CARBOXYL GROUP TO THE N-TERMINAL AMINO GROUP OF THE PEPTIDE LRS BY VIRTUE OF AN AMIDE BOND, AND ALSO COMPRISING THE TARGETING UNIT LRS The targeting agent was synthesized using manual synthesis as de 15 scribed in Example 2 above (analogously to the synthesis in Example Example 1 above) and using the biotinylation procedure described in Example 11 above as the final coupling step. In this final coupling process, D-biotin was employed instead of a protected amino acid. D-biotin was not protected but was em ployed as such. The product was isolated and purified in the manner indicated 20 in Example 2 and identified by positive-mode MALDI-TOF spectroscopy (M+1 ion clearly predominant). MALDI-TOF data (Bio-LRS): calculated molecular mass = 600.31 observed signals: 25 601.34 M+H 623.23 M+Na 639.25 M+K WO 2004/031218 PCT/F12003/000723 53 EXAMPLE 13 SYNTHESIS OF TARGETING AGENT BIO-DLRSK (BIO = D-BIOTIN = VITA MIN H), COMPRISING THE EFFECTOR UNIT D-BIOTIN COUPLED (LINKED DIRECTLY, WITHOUT SPECIFIC LINKER UNITS) VIA ITS CARBOXYL 5 GROUP TO THE N-TERMINAL AMINO GROUP OF THE PEPTIDE DLRSK BY VIRTUE OF AN AMIDE BOND, AND ALSO COMPRISING THE TARGET ING UNIT DLRSK, THAT IS CYCLIC BY VIRTUE AN AMIDE BOND BE TWEEN THE SIDE CHAINS OF D AND K The targeting agent was synthesized using manual synthesis as de 10 scribed in Example 2 above (analogously to the synthesis in Example Example 4 above, including cyclization) and using the biotinylation procedure described in Example 11 above as the final coupling step. In this final coupling process, D-biotin was employed instead of a protected amino acid. D-biotin was not pro tected but was employed as such. The product was isolated and purified in the 15 manner indicated in Example 2 and identified by positive-mode MALDI-TOF spectroscopy (M+1 ion clearly predominant). MALDI-TOF data (Bio-DLRSK cyclic): calculated molecular mass = 825.42 observed signals: 20 826.49 M+H 848.35 M+Na EXAMPLE 14 SYNTHESIS OF TARGETING AGENT BIO-DGRGLRSK (BIO = D-BIOTIN = VITAMIN H), COMPRISING THE EFFECTOR UNIT D-BIOTIN COUPLED 25 (LINKED DIRECTLY, WITHOUT SPECIFIC LINKER UNITS) VIA ITS CAR BOXYL GROUP TO THE N-TERMINAL AMINO GROUP OF THE PEPTIDE DGRGLRSK BY VIRTUE OF AN AMIDE BOND, AND ALSO COMPRISING THE TARGETING UNIT DGRGLRSK, THAT IS CYCLIC BY VIRTUE AN AM IDE BOND BETWEEN THE SIDE CHAINS OF ASPARTIC ACID AND LYSINE 30 The targeting agent was synthesized using manual synthesis as de scribed in Example 2 above (analogously to the synthesis in Example 5 above, including cyclization) and using the biotinylation procedure described in Exam pIe 11 above as the final coupling step. In this final coupling process, D-biotin WO 2004/031218 PCT/F12003/000723 54 was employed instead of a protected amino acid. D-biotin was not protected but was employed as such. The product was isolated and purified in the man ner indicated in Example 2 and identified by positive-mode MALDI-TOF spec troscopy (M+1 ion clearly predominant). 5 MALDI-TOF data (Bio-DGRGLRSK cyclic): calculated molecular mass = 1095.56 observed signal: 1096.51 M+H EXAMPLE 15 10 SYNTHESIS OF TARGETING AGENT BIO-DRGLRSK (BIO = D-BIOTIN = VI TAMIN H), COMPRISING THE EFFECTOR UNIT D-BIOTIN COUPLED (LINKED DIRECTLY, WITHOUT SPECIFIC LINKER UNITS) VIA ITS CAR BOXYL GROUP TO THE N-TERMINAL AMINO GROUP OF THE PEPTIDE DRGLRSK BY VIRTUE OF AN AMIDE BOND, AND ALSO COMPRISING THE 15 TARGETING UNIT DRGLRSK, THAT IS CYCLIC BY VIRTUE AN AMIDE BOND BETWEEN THE SIDE CHAINS OF ASPARTIC ACID AND LYSINE The targeting agent was synthesized using manual synthesis as de scribed in Example 2 above (analogously to the synthesis in Example 6 above, including cyclization) and using the biotinylation procedure described in Exam 20 ple 11 above as the final coupling step. In this final coupling process, D-biotin was employed instead of a protected amino acid. D-biotin was not protected but was employed as such. The product was isolated and purified in the man ner indicated in Example 2 and identified by positive-mode MALDI-TOF spec troscopy (M+1 ion clearly predominant). 25 MALDI-TOF data (Bio-DRGLRSK, cyclic): calculated molecular mass = 1038.56 observed signal: 1039.59 M+H EXAMPLE 16 30 SYNTHESIS OF TARGETING AGENT BIO-DRYYNLRSK (BIO = D-BIOTIN = VITAMIN H), COMPRISING THE EFFECTOR UNIT D-BIOTIN COUPLED (LINKED DIRECTLY, WITHOUT SPECIFIC LINKER UNITS) VIA ITS CAR BOXYL GROUP TO THE N-TERMINAL AMINO GROUP OF THE PEPTIDE DLRSK BY VIRTUE OF AN AMIDE BOND, AND ALSO COMPRISING THE WO 2004/031218 PCT/F12003/000723 55 TARGETING UNIT DRYYNLRSK, THAT IS CYCLIC BY VIRTUE OF AN AM IDE BOND BETWEEN THE SIDE CHAINS OF ASPARTIC ACID AND LYSINE The targeting agent was synthesized using manual synthesis as de scribed in Example 2 above (analogously to the synthesis in Example 7 above, 5 including cyclization) and using the biotinylation procedure described in Exam ple 11 above as the final coupling step. In this final coupling process, D-biotin was employed instead of a protected amino acid. D-biotin was not protected but was employed as such. The product was isolated and purified in the man ner indicated in Example 2 and identified by positive-mode MALDI-TOF spec 10 troscopy (M+1 ion clearly predominant). Mass spectral data (Bio-DRYYNLRSK cyclic): calculated molecular mass = 1421.69 observed signals: 1422.34 M+H 15 EXAMPLE 17 GENERAL METHOD FOR THE CYCLIZATION OF A PEPTIDE AND/OR TARGETING UNIT AND/OR TARGETING AGENT AND/OR TARGETING MOTIF AND/OR TARGETING MOTIF, AND/OR PART THEREOF, IN THE FORM OF A LACTAM (AS MACROLACTAM; BY VIRTUE OF A PEPTIDE 20 BOND BETWEEN LYSINE AND ASPARTIC ACID THOSE WERE INCLUDED IN THE SEQUENCE AT THE ENDS OF AN 'INTERMEDIARY' SEQUENCE) The uncyclized, fully protected, resin-bound peptides were prepared manually by means of the general method described in Example 2 above. Prior to the cyclization, a selective, one-process, dismantling of the 25 side-chain protecting groups of lysine and aspartic acid [the said groups were: 4-methyltrityl on the lysine unit and 2-phenylisopropyl (ester) on the aspartic acid unit] was carried out with diluted TFA (4 % in dichloromethane). The cycli zation involved a condensation between the side-chain carboxyl group of the aspartic acid unit and the 6-amino group (side-chain amino group) of the lysine 30 unit. Activation was by a PyAOP/HOAt/DIPEA reagent mixture (for details and abbreviation explanation, see below) or, alternatively, by the HBTU/HOBt/DIPEA mixture described in Example 2. The equipment, common solvents, and practical techniques were similar to those described in Example 2.

WO 2004/031218 PCT/F12003/000723 56 The initially fully protected resin-bound peptide (0.3 mmol) was shaken under argon atmosphere at room temperature with different solutions (about 10 mL) for the periods of time indicated below, followed by filtration: 1. dichloromethane, for 20 min. 5 2. 4 % (by volume) trifuoroacetic acid in dichloromethane, for 15 min. 3. 0.2 M DIPEA in 1:10 mixture of NMP and dichloromethane, for 3 min. 4. dichloromethane, for 3 min. 5. dichloromethane, for 3 min. 6. dichloromethane, for 3 min. 10 7. DMF, for 3 min. 8. activation, for 4 hours, according to the description below: A mixture of PyAOP and HOAt, or alternatively a mixture of HBTU and HOBt, 3 molecular equivalents of both components with respect to the resin-bound peptide (thus, 0.9 mmol both) in DMF (7 mL), was shaken with the 15 resin for 1 min without filtration, followed by addition of 6 molecular equivalents of 2 M DIPEA in NMP. After step 8 above, the procedures continued as described in Ex ample 2, starting from step 13. The reagents for activation in this type of cyclization were: 20 PyAOP = 7-Azabenzotriazol-1-yloxytris(pyrrolidino)phosphonium hexafluoro phosphate CAS No. 156311-83-0 PE Biosystems Cat. No. GEN076531 Molecular Weight: 521.4 g/mol 25 HOAt = 1-Hydroxy-7-azabenzotriazole 0.5 M solution in DMF Applied Biosystems Cat. No. 4330631 DIPEA = NN-Diisopropylethylamine 2.0 M solution in N-methylpyrrolidinone 30 Applied Biosystems Cat. No. 401517 For materials in the 'HBTU and HOBt' alternative, see the materials indicated in Example 2. Starting materials for the 'special' amino acid units (aspartic acid and lysine), between which the 'extra' peptide bond was formed: 35 WO 2004/031218 PCT/F12003/000723 57 Fmoc-Lys(Mtt) Resin 0.68 mmol/g Bachem Cat. No. D-2565.0005 Fmoc-Asp(2-phenylisopropyl ester)-OH 5 Molecular weight: 473.53 g/mol Bachem Cat. No. B-2475.0005 EXAMPLE 18 SYNTHESIS OF TARGETING AGENT BIO-DSRYNLRSK (BIO = D-BIOTIN = VITAMIN H), COMPRISING THE EFFECTOR UNIT D-BIOTIN COUPLED 10 (LINKED DIRECTLY, WITHOUT SPECIFIC LINKER UNITS) VIA ITS CAR BOXYL GROUP TO THE N-TERMINAL AMINO GROUP OF THE PEPTIDE DLRSK BY VIRTUE OF AN AMIDE BOND, AND ALSO COMPRISING THE TARGETING UNIT DSRYNLRSK, THAT IS CYCLIC BY VIRTUE OF AN AM IDE BOND BETWEEN THE SIDE CHAINS OF ASPARAGINE AND LYSINE 15 The targeting agent was synthesized using manual synthesis as de scribed in Example 2 above (analogously to the synthesis in Example 8 above, including cyclization) and using the biotinylation procedure described in Exam ple 11 above as the final coupling step. In this final coupling process, D-biotin was employed instead of a protected amino acid. D-biotin was not protected 20 but was employed as such. The product was isolated and purified in the man ner indicated in Example 2 and identified by positive-mode MALDI-TOF spec troscopy (M+1 ion clearly predominant). Mass spectral data (Bio-DSRYNLRSK cyclic): calculated molecular mass = 1345.66 25 observed signals: 1346.32 M+H WO 2004/031218 PCT/F12003/000723 58 EXAMPLE 19 SYNTHESIS OF TARGETING AGENT BIO-AHXDLRSK (BIO = D-BIOTIN VITAMIN H), COMPRISING THE EFFECTOR UNIT D-BIOTIN COUPLED (LINKED DIRECTLY, WITHOUT SPECIFIC LINKER UNITS) VIA ITS CAR 5 BOXYL GROUP TO THE N-TERMINAL AMINO GROUP OF THE PEPTIDE AHXDLRSK BY VIRTUE OF AN AMIDE BOND, AND ALSO COMPRISING THE TARGETING UNIT AHXDLRSK, THAT IS CYCLIC BY VIRTUE AN AM IDE BOND BETWEEN THE SIDE CHAINS OF D AND K The targeting agent was synthesized using manual synthesis as de 10 scribed in Example 2 above (analogously to the synthesis in Example 9 above, including cyclization) and using the biotinylation procedure described in Exam ple 11 above as the final coupling step. In this final coupling process, D-biotin was employed instead of a protected amino acid. D-biotin was not protected but was employed as such. The product was isolated and purified in the man 15 ner indicated in Example 2 and identified by positive-mode MALDI-TOF spec troscopy (M+1 ion clearly predominant). MALDI-TOF data (Bio-AhxDLRSK cyclic): calculated molecular mass = 938.50 observed signal: 20 939.50 M+H EXAMPLE 20 SYNTHESIS OF TARGETING AGENT BIO-K-AHXDLRSK (CYCLIC BY VIR TUE OF AN AMIDE BOND BETWEEN ASPARTIC ACID AND TERMINAL LYSINE; BIO = D-BIOTIN = VITAMIN H), COMPRISING ONE EFFECTOR 25 UNIT D-BIOTIN COUPLED (LINKED VIA ONE PLUS ONE LINKER UNITS AND/OR SPACER UNITS AND/OR AS ONE LARGER SPACER AND/OR LINKER UNIT) VIA ITS CARBOXYL GROUP TO THE N-TERMINAL AMINO GROUP OF THE LYSINE RESIDUE (UNIT) AND THIS IN TURN BY VIRTUE OF AN AMIDE BOND TO THE AMINO GROUP OF AHX (6 30 AMINOHEXANOIC ACID) AND THIS BY VIRTUE OF AN AMIDE BOND TO THE AMINO TERMINUS OF THE PEPTIDE DLRSK, AND ALSO COMPRIS ING THE TARGETING UNIT DLRSK The synthesis was carried out as follows: The fully protected resin bound cyclized targeting unit (peptide with spacer/linker unit) AhxDLRSK was WO 2004/031218 PCT/F12003/000723 59 prepared as described in Example 9 above. Next, the sequence AhxDLRSK was continued with one lysine unit (protected with Fmoc-group on N-terminal amino group and with Boc-group on side branch amino group) by means of the general coupling methods described in Example 2. The reagent used as start 5 ing material: Fmoc-L-Lys(tBoc)-OH Finally the still resin-bound and fully protected K-AhxDLRSK was biotinylated according to the general method described in Example 11 . Purifi cation by HPLC gave 30% of the theoretical as overall yield. Identification of 10 the product: positive mode MALDI-TOF mass spectrum: M+1 ion clearly predominant. MALDI-TOF data (Bio-K-AhxDLRSK, cyclic): calculated molecular mass = 1066.60 observed signal: 15 1067.5 M+H EXAMPLE 21 SYNTHESIS OF TARGETING AGENT B104-K3-K-AHXDLRSK (CYCLIC BY VIRTUE OF AN AMIDE BOND BETWEEN ASPARTIC ACID AND TERMINAL LYSINE; BIO = D-BIOTIN = VITAMIN H), COMPRISING FOUR IDENTICAL 20 EFFECTOR UNITS D-BIOTIN COUPLED (LINKED VIA A DENDRIMERIC STRUCTURE THAT CAN BE CONSIDERED AS A COMBINATION OF LINKER UNITS AND/OR SPACER UNITS AND/OR AS ONE LARGER SPACER AND/OR LINKER UNIT) EACH VIA ITS CARBOXYL GROUP TO ONE AMINO GROUP OF A LYSINE RESIDUE (UNIT), EITHER THE N 25 TERMINAL AMINO GROUP OR THE SIDE-CHAIN AMINO GROUP, AND THE DENDRIMERIC STRUCTURE (TWO LYSINES EACH CARRYING TWO EFFECTOR BIOTIN UNITS, THESE LYSINES BEING COUPLED VIA THE CARBOXYL FUNCTIONS TO ONE FURTHER LYSINE AND THIS IN TURN BY VIRTUE OF AN AMIDE BOND TO THE N-TERMINAL AMINO GROUP OF 30 ONE LYSINE (HAVING THE SIDE CHAIN UNCOUPLED) THAT IS SIMI LARLY LINKED TO AHX (6-AMINOHEXANOIC ACID) AND THIS BY VIRTUE OF AN AMIDE BOND TO THE AMINO TERMINUS OF THE PEPTIDE DLRSK, AND ALSO COMPRISING THE TARGETING UNIT DLRSK The product has the formula shown below: 35 WO 2004/031218 PCT/F12003/000723 60 OH

H

2 N H DHO HN O/"' 0 HNN N H H NH H N 0 NH 0 H N H

H

2 N OHN H 0 0 N_( 0 NH N H HN HO0 NH N 0 NH S 0 0 HNN >_NH H S" o H C H and can be stated to comprise a four-fold biotinylated four-branch linker/spacer 5 unit on the amino terminus of K-AhxDLRSK. The synthesis was carried out as follows: The fully protected resin bound, 'on resin' cyclized targeting unit (peptide with two spacer/linker units) K AhxDLRSK was prepared as described in Example 20 above. The branched structure comprising the four biotins and the three lysines was conctructed by 10 means of the general coupling methods described in Example 2, so that the sequence K-AhxDLRSK was continued first with one lysine unit (protected with one Fmoc-group on each of its two amino groups). Then, the procedure (lysine addition) was repeated using doubled amounts of coupling reagents and the doubly protected (Fmoc groups) lysine, in order to couple two more lysine 15 units, one of them on the side-chain amino and one on the amino-terminal WO 2004/031218 PCT/F12003/000723 61 amino group of the first-coupled lysine unit. Reagent used (in addition to the materials described in the referred Examples): Fmoc-L-Lys(Fmoc)-OH Biotinylation was done according to the general method described 5 in Example 11 using 12 molecular equivalents of coupling reagents and biotin, employing the resin-bound branched peptide, to afford a stucture comprising four biotin units. Purification by HPLC gave 44% of the theoretical as overall yield. Identification of the product: 10 positive mode MALDI-TOF mass spectrum: M+1 ion clearly predominant. MALDI-TOF data (Bio4-K3-K-AhxDLRSK, cyclic): calculated molecular mass = 2129.12 observed signal: 2129.89 M+H (the strongest isotopomer is 2130.9) 15 WO 2004/031218 PCT/F12003/000723 62 EXAMPLE 22 SYNTHESIS OF TARGETING AGENT BIO4-K3-K[DTPA]-AHXDLRSK (CY CLIC BY VIRTUE OF AN AMIDE BOND BETWEEN ASPARTIC ACID AND TERMINAL LYSINE; BIO = D-BIOTIN = VITAMIN H; DTPA = DIETHYL 5 ENETRIAMINEPENTAACETIC ACID MINUS ONE OH), COMPRISING TWO TYPES OF EFFECTOR UNITS: FOUR IDENTICAL EFFECTOR UNITS D BIOTIN COUPLED (LINKED VIA A DENDRIMERIC STRUCTURE THAT CAN BE CONSIDERED AS A COMBINATION OF LINKER UNITS AND/OR SPACER UNITS AND/OR AS ONE LARGER SPACER AND/OR LINKER 10 UNIT) EACH VIA ITS CARBOXYL GROUP TO ONE AMINO GROUP OF A LYSINE RESIDUE (UNIT), EITHER THE N-TERMINAL AMINO GROUP OR THE SIDE-CHAIN AMINO GROUP, AND THE DENDRIMERIC STRUCTURE (TWO LYSINES EACH CARRYING TWO EFFECTOR BIOTIN UNITS, THESE LYSINES BEING COUPLED VIA THE CARBOXYL FUNCTIONS TO ONE 15 FURTHER LYSINE AND THIS IN TURN BY VIRTUE OF AN AMIDE BOND TO THE N-TERMINAL AMINO GROUP OF ONE LYSINE (HAVING THE SIDE CHAIN COUPLED VIA AMIDE BOND TO DTPA) THAT IS SIMILARLY LINKED TO AHX (6-AMINOHEXANOIC ACID) AND THIS BY VIRTUE OF AN AMIDE BOND TO THE AMINO TERMINUS OF THE PEPTIDE DLRSK, AND 20 ALSO COMPRISING THE TARGETING UNIT DLRSK The product has the formula shown below: Bio Dtpa 25 Bio-Lys-Lys-Lys-Ahx-Asp-Leu-Arg Bio-Lys CO Ser / \ / Bio HN-Lys 30 and can be stated to comprise a four-fold biotinylated five-branch linker/spacer unit, carrying Dtpa-moiety on one branch, on the N-terminus of peptide AhxDLRSK. The synthesis was carried out as follows: The isolated and purified 35 targeting agent Bio4K3-K-AhxDLRSK was prepared as described in Example WO 2004/031218 PCT/F12003/000723 63 21 above. The product thus obtained was then treated with 10 molecular equivalents of diethylenetriaminepentaacetic dianhydride in DMF solution (0.01 M solution as calculated on the basis of the biotinylated peptide) for 18 hours. After this treatment, the volume was doubled by addition of water to the DMF 5 solution, and the solution was put aside and allowed to stay still for 4 hours. Finally, the solvents were evaporated in vacuo and the residue was mixed in water containing 0.1% trifluoroacetic acid and was filtered and the filtrate was purified by reversed-phase HPLC. The product was identified by its M+1 peak in the MALDI-TOF mass spectrum. 10 Identification of the product: positive mode MALDI-TOF mass spectrum: M+1 ion clearly predominant. MALDI-TOF data (Bio4-K3-K[Dtpa]-AhxDLRSK, cyclic): calculated molecular mass = 2504.24 observed signal: 15 2505.29 M+H EXAMPLE 23 SYNTHESIS OF TARGETING AGENT AOA-DLRSK (AOA = AMINO OXYACETYL = NH20CH2CO), COMPRISING THE EFFECTOR UNIT AMINO-OXYACETIC ACID COUPLED (LINKED DIRECTLY, WITHOUT SPE 20 CIFIC LINKER UNITS) VIA ITS CARBOXYL GROUP TO THE N-TERMINAL AMINO GROUP OF THE PEPTIDE DLRSK BY VIRTUE OF AN AMIDE BOND, AND ALSO COMPRISING THE TARGETING UNIT DLRSK, THAT IS CYCLIC BY VIRTUE LACTAM BRIDGE The targeting agent was synthesized using manual synthesis as de 25 scribed in Example 2 above (analogously to the synthesis in Example Example 4 above, including cyclization). Next, the sequence DLRSK was continued with amino-oxyacetic acid by means of the general coupling methods described in Example 2. Reagent used: 3o Boc-amino-oxyacetic acid; Boc-NH-OCH2-COOH MALDI-TOF data (Aoa-DLRSK, cyclic): calculated molecular mass = 674.37 observed signals: 673.54 M+H WO 2004/031218 PCT/F12003/000723 64 EXAMPLE 24 SYNTHESIS OF TARGETING AGENT CBP-DLRSK [CBP= 5-(1-0 CARBORANYL)-PENTANOYL], COMPRISING THE EFFECTOR UNIT 5-(1-0 CARBORANYL)-PENTANOIC ACID COUPLED (LINKED DIRECTLY, WITH 5 OUT SPECIFIC LINKER UNITS) VIA ITS CARBOXYL GROUP TO THE N TERMINAL AMINO GROUP OF THE PEPTIDE DLRSK BY VIRTUE OF AN AMIDE BOND, AND ALSO COMPRISING THE TARGETING UNIT DLRSK, THAT IS CYCLIC BY VIRTUE LACTAM BRIDGE The targeting agent was synthesized using manual synthesis as de 10 scribed in Example 2 above (analogously to the synthesis in Example Example 4 above, including cyclization). Next, the sequence DLRSK was continued with 5-(1-o-carboranyl)-pentanoic acid by means of the general coupling technique described in Example 2 with the exeption of PyAoP (instead of HBTU) and HOAT (instead of HOBt) and reaction time 4 hours in the treatment step 12 of 15 Example 2. Reagent used: 5-(1-o-carboranyl)-pentanoic acid, Katchem, Prague, Czech Republic, F.W.244.34 g/mol MALDI-TOF data (Cbp-DLRSK, cyclic): 20 calculated molecular mass = 817.60 (basis B10, abund. 20%), 827.57 (basis B1 1 abund. 80%) average molecular weight = 826.01 g/mol observed signals: Multiplet with highest peaks at 826.55 and 827.55: M+H 25 Multiplet with highest peaks at 848.45 and 849.50: M+Na WO 2004/031218 PCT/F12003/000723 65 EXAMPLE 25 SYNTHESIS OF TARGETING AGENT AMF-DLRSK [AMF= 4-AMINO-10 METHYLFOLIC ACYL), COMPRISING THE EFFECTOR UNIT 4-AMINO-10 METHYLFOLIC ACID COUPLED (LINKED DIRECTLY, WITHOUT SPECIFIC 5 LINKER UNITS) VIA ITS CARBOXYL GROUP TO THE N-TERMINAL AMINO GROUP OF THE PEPTIDE DLRSK BY VIRTUE OF AN AMIDE BOND, AND ALSO COMPRISING THE TARGETING UNIT DLRSK, THAT IS CYCLIC BY VIRTUE OF LACTAM BRIDGE The targeting agent was synthesized using manual synthesis as de 10 scribed in Example 2 above (analogously to the synthesis in Example 4 above, including cyclization). Next, the sequence DLRSK was continued with 4-amino 10-methylfolic acid by means of the general coupling technique described in Example 2 with the exceptions of PyAOP (instead of HBTU) and HOAT (in stead of HOBt) and reaction time 5 hours and equimolar ratio of reagents to 15 resin-bound peptide (peptide/PyAOP/HOAT/DIPEA = 1:1:1:2) in the treatment step 12 of Example 2. Reagent used: 4-amino-1 0-methylfolic acid hydrate; (+)amethopterin; methotrexate CAS No. 59-05-2 20 Formula weight: 454.4 g/mol Sigma Cat. No. A-6770 MALDI-TOF data (Amf-DLRSK, cyclic): calculated molecular mass = 1035.50 observed signals: 25 1036.35 M+H WO 2004/031218 PCT/F12003/000723 66 EXAMPLE 26 SYNTHESIS OF TARGETING AGENT DNM-AOA-DLRSK (DNM= DAUNO MYCINE, AOA = AMINO-OXYACETYL = NH20CH2CO), COMPRISING THE EFFECTOR UNIT DAUNOMYCINE COUPLED VIA ITS CARBONYL GROUP 5 BY OXIME LIGATION TO THE AMINO-OXY GROUP OF AOA-DLRSK [A TARGETING AGENT (DERIVATIVE OF PEPTIDE) COMPRISING THE LINKER (LIGATION) UNIT AMINO-OXYACETIC ACID COUPLED (LINKED DIRECTLY, WITHOUT SPECIFIC LINKER UNITS) VIA ITS CARBOXYL GROUP TO THE N-TERMINAL AMINO GROUP OF THE PEPTIDE DLRSK 10 BY VIRTUE OF AN AMIDE BOND], AND ALSO COMPRISING THE TARGET ING UNIT DLRSK, THAT IS CYCLIC BY VIRTUE LACTAM BRIDGE The targeting agent was synthesized by stirring daunomycine with equimolar amount of Aoa-DLRSK described above in Example 23 in methanol solution (concentration 0.0025M) at room temperature in dark for three days. 15 The product was isolated by evaporation of solvents and purified by HPLC. Reagent used: Daunomycin hydrochloride CAS No. 20830-81-3 Molecular weight: 564.0 g/mol 20 ICN Biomedicals, Aurora, Ohio, USA Cat. No. 44583 MALDI-TOF data (Dnm-Aoa-DLRSK, cyclic): calculated molecular mass = 1181.52 observed signal: 25 1182.41 M+H WO 2004/031218 PCT/F12003/000723 67 EXAMPLE 27 SYNTHESIS OF TARGETING AGENT DXRB-AOA-DLRSK (DXRB= DOXORUBICINE, AOA = AMINO-OXYACETYL = NH20CH2CO), COMPRIS ING THE EFFECTOR UNIT DOXORUBICINE COUPLED VIA ITS CARBONYL 5 GROUP BY OXIME LIGATION TO THE AMINO-OXY GROUP OF AOA DLRSK [A TARGETING AGENT (DERIVATIVE OF PEPTIDE) COMPRISING THE LINKER (LIGATION) UNIT AMINO-OXYACETIC ACID COUPLED (LINKED DIRECTLY, WITHOUT SPECIFIC LINKER UNITS) VIA ITS CAR BOXYL GROUP TO THE N-TERMINAL AMINO GROUP OF THE PEPTIDE 10 DLRSK BY VIRTUE OF AN AMIDE BOND], AND ALSO COMPRISING THE TARGETING UNIT DLRSK, THAT IS CYCLIC BY VIRTUE LACTAM BRIDGE The targeting agent was synthesized by stirring doxorubicine with equimolar amount of Aoa-DLRSK described above in Example 23 in methanol solution (concentration 0.0025M) at room temperature in dark for three days. 15 The product was isolated by evaporation of solvents and purified by HPLC. Reagent used: Doxorubicin hydrochloride CAS No. 25316-40-9 Molecular weight: 580.0 g/mol 20 Fluka Cat. No. 44583 MALDI-TOF data (Dxrb-Aoa-DLRSK, cyclic): calculated molecular mass = 1197.52 observed signals: 1198.17 M+H 25 1198.17 M+H EXAMPLE 28 SYNTHESIS OF TARGETING UNIT (PEPTIDE) DLRSGRK THAT IS CYCLIC BY VIRTUE OF LACTAM BRIDGE The functionally protected, resin bound targeting unit (protected 3o peptide), comprising targeting motif LRS, was synthesized by means of man ual synthesis as described in Example 2 above. The following reagents were employed as starting materials (in this order): WO 2004/031218 PCT/F12003/000723 68 Fmoc-Lys(Mtt)-resin Fmoc-L-Arg(Pbf)-OH Fmoc-Gly-OH Fmoc-L-Ser(tBu)-OH 5 Fmoc-L-Arg(Pbf)-OH Fmoc-L-Leu-OH Fmoc-Asp(2-phenylisopropyl ester)-OH After the last cycle of the coupling process,the still resin-bound tar geting unit was subjected to the cyclization process in which an extra amide 10 bond is formed as described in Example 17. After the cyclization process (macrolactam formation) a small sample of the resin (containing the still fully protected cyclized peptide and being suitable starting material for further syn thesis e.g. biotinylation) was subjected to the treatment described in Example 2 for Fmoc removal (steps 1-10 in that Example), after which the sample of 15 peptide was cleaved from the resin by three hours' treatment with the cleavage mixture described in Example 2, and isolated as described in the same Exam ple. Then, the product (cyclic DLRSGRK) was identified with the aid of its positive mode MALDI-TOF mass spectrum, in which the M+1 ion of cyclic 20 DLRSK was clearly predominant. MALDI-TOF data (cyclic DLRSGRK): calculated molecular mass = 812.46 observed signal: 813.69 M+H 25 EXAMPLE 29 SYNTHESIS OF TARGETING AGENT BIO-DLRSGRK (BIO = D-BIOTIN = VI TAMIN H), COMPRISING THE EFFECTOR UNIT D-BIOTIN COUPLED (LINKED DIRECTLY, WITHOUT SPECIFIC LINKER UNITS) VIA ITS CAR BOXYL GROUP TO THE N-TERMINAL AMINO GROUP OF THE PEPTIDE 3o DLRSK BY VIRTUE OF AN AMIDE BOND, AND ALSO COMPRISING THE TARGETING UNIT DLRSGRK, THAT IS CYCLIC BY VIRTUE OF AN AMIDE BOND BETWEEN THE SIDE CHAINS OF ASPARAGINE AND LYSINE The targeting agent was synthesized using manual synthesis as de scribed in Example 2 above (analogously to the synthesis in Example 28 35 above, including cyclization) and using the biotinylation procedure described in WO 2004/031218 PCT/F12003/000723 69 Example 11 above as the final coupling step. In this final coupling process, D biotin was employed instead of a protected amino acid. D-biotin was not pro tected but was employed as such. The product was isolated and purified in the manner indicated in Example 2 and identified by positive-mode MALDI-TOF 5 spectroscopy (M+1 ion clearly predominant). MALDI-TOF data (Bio-DLRSGRK cyclic): calculated molecular mass = 1038.54 observed signals: 1039.74 M+H 10 1061.76 M+Na 1077.60 M+K EXAMPLE 30 SYNTHESIS OF TARGETING UNIT (PEPTIDE) DLRSGRGK. SYNTHESIS OF TARGETING AGENT BIO-DLRSGRGK (BIO = D-BIOTIN = VITAMIN H), 15 COMPRISING THE EFFECTOR UNIT D-BIOTIN COUPLED (LINKED DI RECTLY, WITHOUT SPECIFIC LINKER UNITS) VIA ITS CARBOXYL GROUP TO THE N-TERMINAL AMINO GROUP OF THE PEPTIDE DLRSGRGK BY VIRTUE OF AN AMIDE BOND, AND ALSO COMPRISING THE TARGETING UNIT DLRSGRGK. CYCLIZATION OF THE TARGETING AGENT BIO 20 DLRSGRGK IN THE FORM OF A LACTAM (AS MACROLACTAM) The functionally protected, resin bound targeting unit (protected peptide), comprising targeting motif LRSGRG, was synthesized by means of manual synthesis as described in Example 2 above. The following reagents were employed as starting materials (in this 25 order): Fmoc-Lys(Mtt) Resin Fmoc-Gly-OH Fmoc-L-Arg(Pbf)-OH Fmoc-Gly-OH 30 Fmoc-L-Ser(tBu)-OH Fmoc-L-Arg(Pbf)-OH Fmoc-L-Leu-OH Fmoc-Asp(2-phenylisopropyl ester)-OH The targeting agent was synthesized using the biotinylation proce 35 dure described in Example 11 above as the final coupling step. In this final WO 2004/031218 PCT/F12003/000723 70 coupling process, D-biotin was employed instead of a protected amino acid. D biotin was not protected but was employed as such. The biotinylated, still re sin-bound targeting agent was subjected to the cyclization process in which an extra amide bond is formed as described in Example 17. The product was iso 5 lated and purified in the manner indicated in Example 2 and identified by posi tive-mode MALDI-TOF spectroscopy (M+1 ion clearly predominant). MALDI-TOF data (Bio-DLRSGRGK cyclic): calculated molecular mass = 1065.56 observed signal: 10 1096.55 M+H EXAMPLE 31 IN VIVO TARGETING OF TUMORS IN MICE In this example in vivo targeting of the targeting units prepared in the previous examples is shown for three different types of primary tumors, fi 15 brosarcoma, melanoma and for melanoma metastases in lung. It is shown that the tested targeting units according to the present invention selectively target to primary tumors and to metastases in vivo but not to normal tissues or or gans. CELL LINES AND TUMOR-BEARING MICE 20 The following tumor cell lines were used to produce experimental tumors in mice: "ODC sarcoma cells", (OS), originally derived from tumors that were formed in nude mice to which had been administered NIH3T3 mouse fibro blasts transformed by virtue of ornithine decarboxylase (ODC) overexpression 25 and have been described earlier (Auvinen et al., 1997); A human melanoma cell line C8161 (M) described by Welch et al. (1991). The cell lines were cultured in Dulbecco's Modified Eagle's Medium (DMEM; Bio-Whittaker) supplemented with 5-10% fetal calf serum (FCS; Bio 30 Whittaker), 1% L-glutamine (Bio-Whittaker) and 1% penicillin/streptomycin (Bio-Whittaker).

WO 2004/031218 PCT/F12003/000723 71 EXPERIMENTAL TUMOR PRODUCTION For production of experimental tumors, the cells listed above (OS, KS and melanoma: 0.5 x 106 cells were injected subcutaneously into both flanks of nude mice of the strains Balb/c Ola Hsd-nude, NMRI/nulnu or 5 Athymic-nu (all mice of both strains were from Harlan Laboratories). Tumors were harvested when they had reached a weight of about 0.4 g. Metastases (mostly formed in the lungs) were produced by injection of melanoma cells i.v. into Balb/c Ola Hsd-nude mice. The mice were kept 4-6 weeks, and then targeting experiments were performed. 10 Tumor-bearing or metastase-bearing mice were anesthesized by administering 0.02 mlIg body weight Avertin [10 g 2,2,2-tribromoethanol (Fluka) in 10 ml 2-methyl-2-butanol (Sigma Aldrich)] intraperitoneally (i.p.). IN VIVO TARGETING AND DETECTION OF TARGETING For localization of the targeting peptides OS and melanoma tumor 15 bearing NMRI nude mice were anesthesized and 1 or 2 mg of biotinylated syn thetic targeting peptide (prepared in Example 12) was injected i.v. 5-10 min af ter the i.v. injections, the mice were perfused via the heart using a winged infu sion 25G needle set (Terumo) with 50 ml DMEM. Then, their organs were dis sected and frozen in liquid nitrogen. Tumors and control organs (liver, kidney, 20 spleen, heart, brain) were dissected and frozen in liquid nitrogen. Biotinylated peptides (targeting agents) were detected on 10 mi crometer cryosections using AB (avidin-biotin) -complex containing avidin, and biotinylated HRP (Vectastain ABC-kit, cat no. PK6100; Vector Laboratories) with diaminobenzidine (DAB substrate kit, cat no. 4100, Vector Laboratories). 25 The results of the experiments are shown in Table 2 TABLE 2 Targeting agent dose targeting tumor tumor liver kidney spleen heart brain time type Bio-DLRSK 1mg i.v. 10 min oS + - - - - Bio-DLRSK 1mg Lv. 10 min M + - - - - - WO 2004/031218 PCT/F12003/000723 72 EXAMPLE 32 IN VIVO EFFECT OF TARGETING AGENT COMPRISING CYTOTOXIC EF FECTOR UNIT In this experiment the targeting agent, Dxrb-Aoa-DLRSK prepared 5 in Example 24, comprising a cytotoxic effector unit, doxorubicin, linked by oxime ligation to the cyclic targeting unit DLRSK comprising the targeting motif LRS was used to demonstrate in vivo targeting and therapeutic effect on melanoma tumors. 1 million C8161M/T1 melanoma cells were injected subcutaneously 10 into flank of eight Athymic-nu mice and tumours were allowed to grow for one week. The mice were then divided into three groups those received the follow ing treatments: - Group DMEM: two mice, DMEM only - Group Dox: two mice, 1,43 mg/kg doxorubicin dissolved in DMEM 15 - Group pept + dox: four mice, 1,43 mg/kg Dxrb-Aoa-DLRSK (doxorubicin linked to targeting mofif LRS) dissolved in DMEM (dose equimolar to Group Dox) Treatments were administered i.v. twice a week (Tuesdays and Fri days), total of five doses were injected. Tumours were measured with a calli 20 per in two perpendicular directions on each injection day and on the day the animals were sacrificed. Tumour volume was calculated by the formula for el lipsoid: Volume=(length x width 2 ) x 0.5 The result of the experiment is shown in Figure 1 and confirms that 25 a targeting agent according to the present invention selectively targets to melanoma tumor in vivo and significantly increases the therapeutic effect of doxorubicin. Reagent used: Doxorubicin hydrochloride, CAS No. 25316-40-9, Molecular weight: 580.0 30 g/mol, Fluka Cat. No. 44583 WO 2004/031218 PCT/F12003/000723 73 EXAMPLE 33 GENERAL METHOD FOR THE CYCLIZATION OF A PEPTIDE OR RELATED SUBSTANCE BY VIRTUE OF AN AMIDE BOND BETWEEN D-ORNITHINE AND GLUTAMIC ACID THOSE ARE INCLUDED IN THE SEQUENCE AT THE 5 ENDS OF AN 'INTERMEDIARY' SEQUENCE: FORMATION OF 'HEAD-TO SIDE-CHAIN MACROLACTAM', /E.,'GLU(D-ORN)-RING' The uncyclized, fully protected, resin-bound peptides are prepared manually by means of the general method described above. Prior to the cyclization, a selective, one-process, dismantling of par 10 ticular protecting groups of ornithine and gutamic acid [the said groups are: 2 N-Fmoc on the ornithine unit and 5-(2-trimethylsilylethyl ester) on the glutamic acid unit] is carried out with tetrabutylammonium fluoride solution in DMF. The cyclization involves a condensation between the side-chain carboxyl group of the glutamic acid unit and the 2-amino group (N-terminal amino group) of the 15 ornithine unit. Activation is by a PyAOP/DIPEA reagent mixture (for details and abbreviation explanation, see below) instead of the HBTU/HOBt/DIPEA mix ture described in Example 2. The equipment, common solvents, and practical techniques are similar to those described in Example 2. This method can be modified for lysine (instead of ornithine) and 20 aspartic (instead of glutamic) acid unitst by empoying respective derivatives of those amino acids. The initially fully protected resin-bound peptide (0.3 mmol) is shaken under argon atmosphere at room temperature with different solutions (about 10 mL) for the periods of time indicated below, followed by filtration: 25 1. Dichloromethane, for 20 min. 2. 1 M tetrabutylammonium fluoride in DMF, for 20 min. 3.-5. DMF, for 1 min (three treatments). 6.-8. DCM, for 1 min (three treatments). 9. DMF, for 1 min. 30 10. 0.9 mmol of PyAOP (3 molecular equivalents with respect to the resin-bound peptide) in DMF (7 mL), is shaken with the resin for 1 min without filtration. 11. Addition of 6 molecular equivalents of 2 M DIPEA (thus, 1.8 mmol) in NMP, followed by shaking for 4 hours. 35 After the steps above , the resin is washed etc. as described in the WO 2004/031218 PCT/F12003/000723 74 general procedure for (manual) peptide synthesis (the steps after addition of activated amino acid). The reagent for deprotection prior to cyclization is: Tetrabutylammonium fluoride trihydrate, CAS No. 87749-50-6, molecular 5 weight: 315.51 g/mol, Acros Organics Cat. No. 221080500. The reagents for activation in this type of cyclization are: PyAOP = 7-Azabenzotriazol-1-yloxytris(pyrrolidino)phosphonium hexafluoro phosphate, CAS No. 156311-83-0, PE Biosystems Cat. No. GEN076531, Mo lecular Weight: 521.4 g/mol 10 DIPEA = NN-Diisopropylethylamine, 2.0 M solution in N-methylpyrrolidinone, Applied Biosystems Cat. No. 401517 Starting materials for the 'special' amino acid units (glutamic acid and ornithine), between which the 'extra' amide bond is formed: Fmoc-D-Orn(Mtt)-OH; 2-N-Fmoc-5-N-(4-methyltrityl)-D-ornithine, molecular 15 weight: 610.8 g/mol, Novabiochem Cat. No. 04-13-1012. Fmoc-L-Glu(OTMSEt)-ONa; N-2-Fmoc-glutamic acid 5-(2-trimethylsilylethyl) ester sodium salt, molecular weight: 468.60 g/mol, Novabiochem Cat. No. 04 12-1231. EXAMPLE 34 20 SYNTHESIS OF TARGETING UNIT (PEPTIDE) D-ORNLRSE-AMIDE, CY CLIC BY VIRTUE OF AN AMIDE BOND BETWEEN THE SIDE CHAIN OF GLUTAMIC ACID UNIT AND THE a-AMINO GROUP OF D-ORNITHINE The functionally protected, resin bound targeting unit (protected peptide), comprising targeting motif LRS, was synthesized by means of man 25 ual synthesis as described in Example 2 above, in which the the "empty" resin was deprotected prior to the first coupling in the same manner as described for the the pre-loaded resins (steps 1-11 in Example 2). The following reagents were employed as starting materials (in this order): 30 Rink amide MBHA Resin Fmoc-L-Glu(OTMSEt)-OH Fmoc-L-Ser(tBu)-OH Fmoc-L-Arg(Pbf)-OH Fmoc-L-Leu-OH 35 Fmoc-D-Orn(Mtt)-OH WO 2004/031218 PCT/F12003/000723 75 After the last cycle of the coupling process, the still resin-bound tar geting unit was subjected to the cyclization process in which an extra amide bond is formed as described in Example 33. Next, a sample of peptide was cleaved from the resin by three hours' treatment with the cleavage mixture de 5 scribed in Example 2, and isolated as described in the same example. Then, the product was identified with the aid of its positive mode MALDI-TOF mass spectrum, in which the M+1 ion of cyclic D-OrnLRSE-amide was clearly predominant. MALDI-TOF data (cyclic D-OrnLRSE-NH 2 ): 10 calculated molecular mass = 598.36 observed signal: 599.42 M+1 EXAMPLE 35 SYNTHESIS OF TARGETING UNIT (PEPTIDE) D-ORNLRSGRGEG, CYCLIC 15 BY VIRTUE OF AN AMIDE BOND BETWEEN THE SIDE CHAIN OF GLU TAMIC ACID UNIT AND THE a-AMINO GROUP OF D-ORNITHINE The functionally protected, resin bound targeting unit (protected peptide), was synthesized by means of manual synthesis as described in Ex ample 2 above. 20 The following reagents were employed as starting materials (in this order): Fmoc-Gly Resin Fmoc-L-Glu(OTMSEt)OH Fmoc-Gly-OH 25 Fmoc-L-Arg(Pbf)-OH Fmoc-Gly-OH Fmoc-L-Ser(tBu)-OH Fmoc-L-Arg(Pbf)-OH Fmoc-L-Leu-OH 30 Fmoc-D-Orn(Mtt)-OH After the last cycle of the coupling process, the still resin-bound tar geting unit was subjected to the cyclization process in which an extra amide bond is formed as described in Example 33. Next a sample of peptide was cleaved from the resin by three hours' treatment with the cleavage mixture de 35 scribed in Example 2, and isolated as described in the same example.

WO 2004/031218 PCT/F12003/000723 76 Then, the product was identified with the aid of its positive mode MALDI-TOF mass spectrum by means of the M+1 ion of cyclic D OrnLRSGRGEG. MALDI-TOF data (cyclic D-OrnLRSGRGEG): 5 calculated molecular mass = 926,50 observed signal: 927.45 M+1 EXAMPLE 36 SYNTHESIS OF TARGETING AGENT D-ORN(DOTA)LRSE-AMIDE (DOTA= 10 1,4,7,10-TETRAAZACYCLODODECANE-1,4,7,10-TETRAACETIC ACID COUPLED BY ITS ONE CARBOXYL) , CYCLIC BY VIRTUE OF AN AMIDE BOND BETWEEN THE SIDE CHAIN OF GLUTAMIC ACID UNIT AND THE a AMINO GROUP OF D-ORNITHINE

HO

2 C O Q NI NH N N C HN HN NH 2 N H HN Z H N O_ NH O

CO

2 H ' N HO

H

2 N o 15 The functionally protected, resin-bound, and cyclized targeting unit, comprising the targeting motif LRS, was synthesized by means of manual syn thesis as described in Example 34 above. Next, the resin was treated with di luted TFA (4 % in dichloromethane) in the manner described in example 17 (steps 1-7) to cleave the side chain protecting Mtt-group of ornithine. The still 20 resin-bound unit was then coupled with DOTA-tris-tert-butyl ester by means of the general method described in Example 2 (steps 12-18) using HBTU/HOBt/DIPEA activation. Reagent used: DOTA-tris(tBu ester). The product was cleaved and isolated as described in Example 2 25 and identified with the aid of its positive mode MALDI-TOF mass spectrum, in WO 2004/031218 PCT/F12003/000723 77 which the M+1 ion of cyclic D-Orn(Dota)LRSE-amide was clearly predomi nant. MALDI-TOF data [cyclic D-Orn(Dota)LRSE-NH 2 ]: calculated molecular mass = 984.54 5 observed signal: 985.52 M+1 EXAMPLE 37 SYNTHESIS OF TARGETING UNIT (PEPTIDE) KLRSD-AMIDE, CYCLIC BY VIRTUE OF AN AMIDE BOND BETWEEN THE SIDE CHAIN OF ASPARTIC 10 ACID UNIT AND THE a-AMINO GROUP OF LYSINE The functionally protected, resin bound targeting unit (protected peptide), comprising the targeting motif LRS, was synthesized by means of manual synthesis as described in Example 2 above, in which the the "empty" resin was deprotected prior to the first coupling in the same manner as de 15 scribed for the the pre-loaded resins (steps 1-11 in Example 2). The following reagents were employed as starting materials (in this order): Rink amide MBHA Resin Fmoc-L-Asp(OTMSEt)-OH 20 Fmoc-L-Ser(tBu)-OH Fmoc-L-Arg(Pbf)-OH Fmoc-L-Leu-OH Fmoc-L-Lys(Mtt)-OH After the last cycle of the coupling process, the still resin-bound tar 25 geting unit was subjected to the cyclization process in which an extra amide bond is formed as described in Example 33 (as modification which replaces Glu with Asp and Lys with Orn). Next, a sample of peptide was cleaved from the resin by three hours' treatment with the cleavage mixture described in Ex ample 2, and isolated as described in the same example. 30 Then, the product was identified with the aid of its positive mode MALDI-TOF mass spectrum by means of M+1 ion. MALDI-TOF data (cyclic KLRSD-NH 2 ): calculated molecular mass = 598.36 observed signal: 35 599.21 M+1 WO 2004/031218 PCT/F12003/000723 78 EXAMPLE 38 SYNTHESIS OF TARGETING AGENT K(DOTA)LRSD-AMIDE (DOTA= 1,4,7,1 0-TETRAAZACYCLODODECANE-1,4,7,1 0-TETRAACETIC ACID COUPLED BY ITS ONE CARBOXYL), CYCLIC BY VIRTUE OF AN AMIDE 5 BOND BETWEEN THE SIDE CHAIN OF ASPARIC ACID UNIT AND THE a AMINO GROUP OF LYSINE The functionally protected, resin-bound, and cyclized targeting unit, comprising the targeting motif LRS, was synthesized by means of manual syn thesis as described in Example 37 above. Next, the resin was treated with di 10 luted TFA (4% in dichloromethane) in the manner described in example 17 (steps 1-7) to cleave the side chain protecting Mtt-group of lysine. The still resin-bound unit was then coupled with DOTA-tris-tert-butyl ester by means of the general method described in Example 2 (steps 12-18) using HBTU/HOBt/DIPEA activation. 15 Reagent used: DOTA-tris(tBu ester). The product was cleaved and isolated as described in Example 2 and identified with the aid of its positive mode MALDI-TOF mass spectrum by means of M+1 ion. 20 MALDI-TOF data [cyclic K(Dota)-LRSE-NH 2 ]: calculated molecular mass = 984.54 observed signal: 985.52 M+1 EXAMPLE 39 25 SYNTHESIS OF TARGETING UNIT AC-DLRSK-AHX, CYCLIC VIA SIDE CHAINS OF ASPARTIC ACID AND LYSINE The preparation of Ac-DLRSK-Ahx was executed by manual solid phase peptide synthesis technique that is described in details in Example 2. The binding of the first structural component (moiety), 6-amino 30 hexanoic acid (= Ahx) whose amino function was protected by 9-fluorenyl methyloxycarbonyl group (= Fmoc group), to a hydroxyl-functionalized peptide synthesis resin was carried out by means of dichlorobenzoyl chloride method WO 2004/031218 PCT/F12003/000723 79 (the "equivalents" below are molecular or "mol" amounts relative to the loading capacity of the resin): The unloaded ("empty") resin was first washed by shaking with N,N-dimethylformamide (= DMF) for 20 min and filtered. After addition of five 5 equivalents of the Fmoc-protected 6-aminohexanoic acid (Fmoc-Ahx-OH) in DMF (0.2 M solution) and eight equivalents of pyridine onto the resin it was shaked for 3 min. Next, five equivalents of 2,6-dichlorobenzoychloride was added and the mixture was shaken for 18 h (overnight). After the lengthy treatment the resin was filtered and washed sev 10 eral times with DMF and dichloromethane in the way described in Example 2 (steps 13 -18). Next, the resin was shaken for 2 hours with a mixture of acetic anhydride (2M solution, 94 equivalents) and NN-diisopropylethylamine (DIPEA, 1.6 M solution, 80 equivalents) in N-methyl pyrrolidinone (NMP) solu tion, filtered and washed like earlier ending up in drying at argon gas flow. 15 The reagents used this far were: HMP Resin, loading capacity: 1.16 mmol/g, Applied Biosystems Cat. No. 400957. 2,6-dichlorobenzoyl chloride, CAS No. 225-102-4, molecular weight: 209.46 g/mol, Lancaster (Morecambe, England), Cat. No. 8922. 20 Pyridine, Merck Art. No. 9728. Fmoc-6-aminohexanoic acid (Fmoc-Ahx-OH) , CAS No. 88574-06-5, Novabio chem Cat. No. 04-12-1111, Molecular Weight: 353.4 g/mol. Acetic anhydride, Fuka Cat. No. 45830. From this on, the synthesis proceeds according to the general 25 method decribed in Example 2. The stuctural reagents used next in this syn thesis, are in sequence as follows: Fmoc-Lys(Mtt)-OH Fmoc-L-Ser(tBu)-OH Fmoc-L-Arg(Pbf)-OH 30 Fmoc-L-Leu-OH Fmoc-Asp(2-phenylisopropyl ester)-OH The still resin-bound product was next cyclized according to Exam ple 17. Finally the sequence was continued with acetic acid (i.e. end-capped at amino terminal) as follows: Amino protecting Fmoc-group was removed as de 35 scribed in Example 2 (steps 1-10). Then the still resin-bound product was treated with the mixture of acetic anhydride and DIPEA in NMP like was done WO 2004/031218 PCT/F12003/000723 80 after the initial binding of Ahx moiety to the resin. In the end the product was released from the resin and purified as described in Example 2. Identification was based on M+1 ion of MALDI mass spectrum. MALDI-TOF data (cyclic Ac-DLRSK-Ahx) 5 calculated molecular mass = 754.43 observed signal: 755.60 EXAMPLE 40 SYNTHESIS OF TARGETING AGENT AC-DLRSK-AHX-DOX (DOX = DOXORUBICIN COUPLED VIA ITS AMINO GROUP) COMPRICING 10 DOXORUBICIN LINKED VIA AN AMIDE BOND TO THE CARBOXYL GROUP OF C-TERMINAL SPACER MIOETY (AHX = 6-AMINOHEXANOYL) OF THE N-CAPPED (AC = ACETYL) CYCLIC TARGETING UNIT AC-DLRSK-AHX COMPRICING TARGETING MOTIF LRS The "targeting unit" compound ( a peptide derivative) Ac-DLRSK 15 Ahx was prepared as described in Example 39. Doxorubicine was linked to pu rified Ac-DLRSK-Ahx in NN-dimethylformamide (= DMF) solution by means of PyAOP/DIPEA activation as follows: Equimolar amounts of Ac-DLRSK-Ahx and PyAOP were combined in DMF as 0.05 M solution, two molar equivalents of DIPEA (2 M solution in 20 NMP) was mixed in and after five minutes equimolar (in respect to Ac-DLRSK Ahx) amount of doxorubicin hydrochloride (0.05 M solution in DMF) was added. After the reaction was allowed to proceed one hour at dark (protected from light) the mixture was diluted with diethyl ether. The centrifugued solid precipitate was purified by reverse phase HPLC chromatography and identified 25 by positive mode MALDI mass spectrum as described in Example 2.

WO 2004/031218 PCT/F12003/000723 81 Formula of Ac-DLRSK-Ahx-Dox: O1- NH 0 N NH O HO NH O OH N HN NH NH HO OH N HO NH OH O 0 OMe 5 Material used: Doxorubicin hydrochloride, CAS No. 25316-40-9, molecular weight: 580.0 g/mol, Sigma Cat. No. D-1515. MALDI-TOF data (Ac-DLRSK-Ahx-Dox, -DLRSK- moiety cyclic): calculated molecular mass = 1279.60 10 observed signal: 1280.29 M+1 EXAMPLE 41 SYNTHESIS OF TARGETING AGENT AMF-AHXDLRSK [AMF = 4-AMINO 10-METHYLFOLIC ACYL], COMPRISING THE EFFECTOR UNIT 4-AMINO 15 10-METHYLFOLIC ACID COUPLED VIA ITS CARBOXYL GROUP TO THE N TERMINAL AMINO GROUP OF THE PEPTIDE AHXDLRSK BY VIRTUE OF AN AMIDE BOND, AND ALSO COMPRISING THE TARGETING UNIT DLRSK, THAT IS CYCLIC BY VIRTUE OF LACTAM BRIDGE BETWEEN THE SIDE CHAINS OF THE OUTERMOST MEMBERS OF THE SEQUENCE 20 The resin-bound, a spacer (= Ahx) comprising targeting unit (AhxDLRSK) was prepared as described in Example 9 above, including cycli zation (according to Example 21). Next, the sequence AhxDLRSK was contin ued on resin with glutamic acid by means of the general coupling technique WO 2004/031218 PCT/F12003/000723 82 described in Example 2. As final coupling the sequence (now E-AhxDLRSK, where "E" will be a part of the "Amf" moiety) was ended with "Amf minus E acid", i.e. 4-(N-(2,4-diamino-6-pteridinyl-methyl)-N-methylamino]-benzoic acid hemihydrochloride dihydrate, by means of the general coupling techniques 5 with the exceptions of PyAOP as activation reagent (instead of HBTU and HOBt), reaction time 5 hours, and nealy equimolar ratio of reagents to resin bound peptide in the treatment step 12 of Example 2. The stoichiometric reagent ratios in that step were: "resin-bound peptide" / "Amf minus E acid" / PyAOP / DIPEA = 1 : 1.2 : 1.2 10 2.4, (time 5 h). After isolation and purification, according to Example 2, the product was identified on the basis of M+1 ion in positive mode MALDI mass spectrum. Reagents used: Fmoc-L-Glu(OtBu)-OH, CAS No. 71989-18-9, Applied Biosystems Cat. No. 15 GEN911036, Molecular Weight: 425.5 g/mol. 4-[N-(2,4-diamino-6-pteridinyl-methyl)-N-methylamino]-benzoic acid hemihy drochioride dihydrate, CAS No. 19741-14-1, Aldrich Cat No. 86,155-3, molecu lar weight: 379.59 g/mol, designated as "Amf minus E acid". MALDI-TOF data (Amf-AhxDLRSK, cyclic): 20 calculated molecular mass = 1148.58 observed signals: 1149.62 M+H WO 2004/031218 PCT/F12003/000723 83 EXAMPLE 42 SYNTHESIS OF TARGETING AGENT PTXSUC-AHXDLRSK (PTXSUC = PACLITAXEL MONOSUCCINATE), COMPRISING THE EFFECTOR UNIT PACLITAXEL AS MONOSUCCINATE COUPLED VIA ITS SUCCINYL (SUC 5 CINIC CARBOXYL) GROUP TO THE AMINO GROUP AT 6-AMINO HEXANOYL (= AHX) MOIETY OF THE PEPTIDE AHXDLRSK BY VIRTUE OF AN AMIDE BOND (OR TARGETING AGENT WHERE EFFECTOR UNIT PA CLITAXEL IS LINKED VIA THE SPACER UNIT 6-(SUCCINYLAMINO) HEXANOYL TO THE TARGETING UNIT DLRSK), AND ALSO COMPRISING 10 THE TARGETING UNIT DLRSK, THAT IS CYCLIC BY VIRTUE OF AN AM IDE BOND BETWEEN THE SIDE CHAINS OF THE OUTERMOST MEMBERS OF THE SEQUENCE 0 H 0

H

2 N HH H O N NH - , _N NH N o,, HN N o O O H 0 O HO 0 0N 0Ph O OH HO 0 15 Paclitaxel succinate, described in the end of this example, was dis solved as 0.012 M solution in DMF and equimolar amount of 0.05 M PyAOP in DMF was added, followed by double molar amount of 2.0 M DIPEA in NMP. After 2 minutes equimolar amount (per paclitaxel succinate) of side-chain-to side-chain cyclic targeting compound AhxDLRSK, described in Example 9 20 above, was added as 0.015 M solution in DMF. After staying overnight the mix ture was diluted with diethyl ether. The centrifuged solid precipitate was puri fied by reverse phase HPLC chromatography as described in Example2, in cluding the identification of the product based on its M+1 ion in the positive mode MALDI-TOF mass spectrum. 25 WO 2004/031218 PCT/F12003/000723 84 MALDI-TOF data (PtxSuc-AhxDLRSK, cyclic): calculated molecular mass = 1647.76 observed signal: 1648.57 M+1 5 Packlitaxel succinate was synthesized by following the procedure described in the article: Chun-Ming Huang, Ying-Ta Wu and Shui-Tein Chen (2000). Targeting delivery of paclitaxel into tumor cells via somatostatin recep tor endocytosis. Chemistry & Biology 2000, Vol 7 No 7. 453-461. Herewith 0.05 M paclitaxel in pyridine was stirred with 12-fold ex 10 cess of succinic anhydride for 3 hours. After evaporation of the solvent in re duced pressure, the residue was dissolved in water and freeze-dryed (lyophi lized). Materials used in the synthesis of paclitaxel succinate: Paclitaxel (from Taxus yannesis), CAS No. 33069-62-4, molecular weight: 15 853.9 g/mol, Sigma product No. T-1912. Succinic anhydride, CAS No. 108-30-5, molecular weight: 100.08 g/mol, Fuka product No. 14089.

WO 2004/031218 PCT/F12003/000723 85 EXAMPLE 43 SYNTHESIS OF TARGETING AGENT CPTC-AHXDLRSK [CPTC = (S)-(+) CAMPTOTHECIN LINKED AS ESTER AT ITS HYDROXYL GROUP VIA CARBONIC ACYL, I.E (S)-(+)-CAMPTOTHECIN CARBONYL MOIETY], 5 COMPRISING THE EFFECTOR UNIT CAMPTOTHECIN CARBONATE COU PLED TO THE AMINO GROUP AT 6-AMINOHEXANOYL (= AHX) MOIETY OF THE PEPTIDE AHXDLRSK BY VIRTUE OF AN AMIDE BOND (OR TAR GETING AGENT WHERE EFFECTOR UNIT (S)-(+)-CAMPTOTHECIN IS LINKED VIA THE SPACER UNIT 6-(CARBONYLAMINO)-HEXANOYL TO 10 THE TARGETING UNIT DLRSK), AND ALSO COMPRISING THE TARGET ING UNIT DLRSK, THAT IS CYCLIC BY VIRTUE OF AN AMIDE BOND BE TWEEN THE SIDE CHAINS OF THE OUTERMOST MEMBERS OF THE SE QUENCE 0 HH 2 N O O - NH '-NH 0 0 ~ H 0 0 N O N H H N N NH 0 OH HO 15 Camptothecin p-nitrophenylcarbonate, described in the end of this example, was dissolved as 0.02 M solution in DMF and combined with 0.04 M solution of equimolar amount of cyclic targeting compound AhxDLRSK, de scribed in Example 9 above, in the same solvent. After staying overnight, 2 M 20 DIPEA in NMP was added in 10% excess (i.e. equimolar amount multiplied by 1.1). After being stirred overnight the mixture was diluted with diethyl ether and the centrifuged solid precipitate was purified by reverse phase HPLC chroma tography as described in Example2, including the identification of the product based on its M+1 ion in the positive mode MALDI-TOF mass spectrum. 25 WO 2004/031218 PCT/F12003/000723 86 MALDI-TOF data (Cptc-AhxDLRSK, cyclic): Calculated molecular mass = 1086.51 Observed signal: 1087.26 M+1 5 The synthesis of camptothecin p-nitrophenylcarbonate: 0.29 mmol of 4-nitrophenyl chloroformate and 0.10 mmol of (S)-(+)-camptothecin were dissolved in 12 mL of dichloromethane (DCM). Next, 1.71 mmol of 4 (dimethylamino)-pyridine (DMAP) was added to the DCM solution on cooling water bath. The mixture was then shaken for two hours followed by dilution 10 with 30 mL of DCM. After washings: twice with 0.1% hydrochloric acid and once with saturated aqueous sodium chloride solution, the DCM solution was dried with disodium sulfate, filtered, and concentrated to small volume. The product was precipitated by addition of diethyl ether and gathered after cen trifugation. 15 Materials used in the synthesis of camptothecin p-nitrophenyl carbonate: 4-nitrophenyl chloroformate, CAS No. 7693-46-1, molecular weight: 201.57 g/mol, Fluka product No. 23240. (S)-(+)-camptothecin, CAS No. 7689-03-4, molecular weight: 348.36 g/mol, Al 20 drich product No. 36,563-7. DMAP; 4-(dimethylamino)-pyridine, CAS No. 1122-58-3, molecular weight 122.17, Fluka product No. 29224. EXAMPLE 44 SYNTHESIS OF TARGETING AGENT DOTA-AHXDLRSGRGK-AMIDE 25 (DOTA = 1,4,7,10-TETRAAZACYCLODODECANE-1,4,7,10-TETRAACETIC ACID COUPLED BY ITS ONE CARBOXYL), COMPRISING POTENTIAL METAL CHELATOR GROUP "DOTA"AS EFFECTOR LINKED VIA A SPACER UNIT, 6-AMINOHEXANOYL (= AHX), TO THE TARGETING UNIT "DLRSGRGK", THAT IN TURN, IS CYCLIC VIA THE SIDE CHAINS OF AS 30 PARTIC ACID MOIETY ("D") AND LYSINE ("K"). ALL THE BONDS NEEDED FOR MENTIONED COMBINATION ARE AMIDE BONDS The functionally protected, resin bound targeting unit (protected peptide), comprising the targeting motif LRS, was synthesized by means of manual synthesis as described in Example 2 above, in which the the "empty" WO 2004/031218 PCT/F12003/000723 87 resin was deprotected prior to the first coupling in the same manner as de scribed for the the pre-loaded resins (steps 1-11 in Example 2). The following reagents were employed as starting materials (in this order): 5 Rink amide MBHA Resin Fmoc-L-Lys(Mtt)-OH Fmoc-Gly-OH Fmoc-L-Arg(Pbf)-OH Fmoc-Gly-OH 10 Fmoc-L-Ser(tBu)-OH Fmoc-L-Arg(Pbf)-OH Fmoc-L-Leu-OH Fmoc-L-Asp(2-Phenylisopropyl ester)-OH After these coupling cycles the still resin-bound targeting unit was 15 subjected to the cyclization process in which an extra amide bond is formed as described in Example 17. Next, still two couping cycles were carried out as de scribed in Example 2 above employing following starting materials: Fmoc-6-aminohexanoic acid DOTA-tris-(tBu ester). 20 Finally the the resin was cleaved and the product purified as de scribed in Example 2, and identified by means of its M+1 ion in positive mode MALDI-TOF mass spectrum. MALDI-TOF data (Dota-AhxDLRSGRGK-amide): calculated molecular mass = 1367.76 25 observed signal: 1368.70 M+1 WO 2004/031218 PCT/F12003/000723 88 EXAMPLE 45 SYNTHESIS OF TARGETING AGENT GD-DOTA-AHXDLRSGRGK-AMIDE (GD = CHELATED GADOLINIUM"', DOTA = 1,4,7,10 TETRAAZACYCLODODECANE-1,4,7,10-TETRAACETIC ACID MINUS 5 THREE HYDROGENS COUPLED BY ITS ONE ACYL TO FORM AN AMIDE BOND), COMPRISING GADOLINIUM ATOM AS DETECTABLE EFFECTOR LINKED BY CHELATOR "DOTA"VIA A SPACER UNIT, 6-AMINOHEXANOYL (= AHX), TO THE TARGETING UNIT "DLRSGRGK", THAT IN TURN, IS CY CLIC VIA THE SIDE CHAINS OF ASPARTIC ACID MOIETY ("D") AND LY 10 SINE ("K") 14.9 mg of the chelator-peptide compound described in Example 44 and 24 mg of Gadolinium trichloride hydrate (five-fold excess) was dissolved in 1 mL of aqueous 0.05 M ammonium acetate. Next day the solution was sub jected to reverse phase HPLC purification as described in Example 2 with the 15 exception of 0.05 M ammonium acetate as aqueous buffer (instead of aqueous TFA). Identification was positive mode MALDI-TOF mass spectrum at neutral matrix (no TFA). The purified yield was 9.4 mg. MALDI-TOF data: Calculated molecular mass: 1522.66 based on the most abundant isotopes 20 Observed signal M+1: 1523.66 as typical isotopic pattern LIST OF REAGENTS 4-Amino-10 methylfolic acid; (+)amethopterin; methotrexate hydrate; Formula weight: 454.4 g/mol, CAS No. 59-05-2, Sigma A-6770 Boc-amino-oxyacetic acid; Boc-NH-OCH2-COOH, Molecular weight: 191.2 25 g/mol, CAS No., Novabiochem Cat. No. 01-63-0060 Boc-Cys (Trt)-OH, CAS No: 21947-98-8, Novabiochem, cat. no 04-12-0020 D-Biotin (Vitamin H), CAS No. 58-85-5, molecular weight: 244.3, Sigma B 4501, 99% DL-2,3-diaminopropionic acid monohydrochloride, CAS No. 54897-59-5 30 C3H8N202.HCI, Acros Organics, New Jersey USA; Ceel Belgium, Cat. No. 204670050 Diethylenetriaminepentaacetic dianhydride, CAS No. 23911-26-4 molecular weight: 357.32, Aldrich cat. no. 28,402-5 Fmoc-6-aminohexanoic acid (Fmoc-6-Ahx-OH), CAS No. 88574-06-5 WO 2004/031218 PCT/F12003/000723 89 Fmoc-Asp(2-phenylisopropyl ester)-OH, Molecular weight: 473.53 g/mol Bachem Cat. No. B-2475.0005 Fmoc-L-Asn-OH, CAS No. 71089-16-7, Applied Biosystems, cat. no: GEN 911018 5 Fmoc-Gly Resin, Applied Biosystems Cat. No. 401421, 0.65 mmol/g Fmoc-Gly-OH, CAS No. 29022-11-5, Novabiochem Cat. No. 04-12-1001 Molecular Weight: 297.3 g/mol Fmoc-L-Asn-OH, Applied Biosystems Cat. No. Gen 911018, Molecular weight: 354.40 10 Fmoc-L-Arg(Pbf)-OH, CAS No. 154445-77-9, Applied Biosystems Cat. No. GEN911097, Molecular Weight: 648.8 g/mol Fmoc-L-Cys(Trt)-OH, CAS No. 103213-32-7, Applied Biosystems Cat. No. GEN911027, Molecular Weight: 585.7 g/mol Fmoc-L-Leu-OH, CAS No. 35661-60-0, Applied Biosystems Cat. No. 15 GEN911048, Molecular Weight: 353.4 g/mol Fmoc-L-Lys(Fmoc)-OH, CAS No. 78081-87-5, Molecular weight: 590.7 g/mol PerSeptive Biosystems Cat. No. GEN911095, Hamburg, Germany Fmoc-Lys(Mtt) Resin, 0.68 mmol/g, Bachem Cat. No. D-2565.0005 Fmoc-L-Lys(tBoc)-OH, CAS No. 71989-26-9, Molecular Weight: 468.6 g/mol 20 Applied Biosystems Cat. No. GEN911051 Fmoc-Ser(tBu) Resin, Applied Biosystems Cat. No. 401429, 0.64 mmol/g Fmoc-L-Ser(tBu)-OH, CAS No. 71989-33-8, Perseptive Biosystems Cat. No. GEN911062, Molecular Weight: 383.4 g/mol Fmoc-L-Tyr(tBu)-OH, Applied Biosystems Cat. No. GEN911068 25 CAS No. 71989-38-3, M.W. 459.5 LIST OF SUPPLIERS Applied Biosystems, Warrington, WA1 4SR,United Kingdom Bachem AG, Hauptstrasse 144, CH-4416 Bubendorf, Switzerland Calbiochem-Novabiochem, CH-4448 Laufelfingen, Switzerland 30 Fluka Chemie GmbH, Buchs, Switzerland Merck KGaA, Darmstadt, Germany PE Biosystems, Warrington, United Kingdom Perseptive Biosystems, Warrington, United Kingdom/HamburgGermany Sigma Aldrich Chemie, Steinheim Germany 35 (also Riedel-deHa~n) Sigma-Genosys LTD, Pampisford, Cambridge, UK WO 2004/031218 PCT/F12003/000723 90 Bio-Whittaker, Verviers, Belgium Harlan Laboratories, Horst, The Netherlands Genset SA, Paris, France AmershamPharmacia Biotech, Uppsala, Sweden 5 Qiagen, Hilden, Germany Terumo, Leuven, Belgium Vector Laboratories, Burlingame, USA WO 2004/031218 PCT/F12003/000723 91 REFERENCES Adams, GP, Schier R. Generating improved single-chain Fv molecules for tu mor targeting. J Immunol Methods 1999;231:249-60. Arap, W., Pasqualini, R., and Ruoslahti, E. (1998). Chemotherapy targeted to 5 tumor vasculature. Curr. Op. Oncol. 10: 560-565. Auvinen, M, Laine, A., Paasinen-Sohns, A., Kangas, A., Saksela, 0., Anders son, L. C., and Hbltts, E. (1997). Human ornithine decarboxylase overproducing NIH3T3 cells induce rapidly growing, highly vascularized tumors in nude mice. Cancer Res. 57: 3016-25. 10 Bachem AG, SASRINTM , A review of its manyfold applications including many useful procedures, complied by Mergler M., 2 nd revised and enlarged edition 1999, 1999 by BACHEM AG, CH-4416 Bubendorf, Switzerland. Bachem 2001, Peptides and Biochemicals, Immunochemicals, The New 2001 BACHEM catalog, BACHEM AG, Hauptstrasse 144, CH-4416 Bubendorf 15 Switzerland. Beckman, G.; Beckman, L.; Ponten J. and Westermark B. (1971) G-6-PD and PGM phenotypes of 16 continuous human tumor cell lines. Evidence against cross-contamination and contamination by HeLa cells. Hum. Hered. 21: 238 241. 20 Biosystems Solutions, Issue 2 - September 2001, p. 30. AB Applied Biosys tems. Chan, W.C. Bycroft, B.W., Evans, D.J. and White, P.D. A novel 4-aminobenzyl ester-based carboxy-protecting group for synthesis of atypical peptides by Fmoc-But solid-phase chemistry. (1995) J. Chem. Soc., Chem. Commun., 25 1995, p. 2209. Ellerby, H. M., Arap, W., Ellerby, L. M., Kain, R., Andrusiak, R., Rio, G. D., Kra jewski, S., Lombardo, C. R., Rao, R., Ruoslahti, E., Bredesen, D. E., and Pas qualini, R. 1(1999). Anti-cancer activity of targeted pro-apoptotic peptides. Nat. Med. 9: 1032-1038. 30 Fluka Chemika, Peptide and Peptidomimetic Synthesis, Reagents for Drug Discovery, 2000 Fluka Chemie GmbH, Buchs, Fluka, Speciality Chemicals and Analytical Reagents. Fogh J.; Fogh JM. and Orfeo T. (1977) One hundred and twenty-seven cu tured human tumor cell lines producing tumors in nude mice. J. Nat. Cancer 35 Inst. 59: 221-226.

WO 2004/031218 PCT/F12003/000723 92 Herndier BG, Werner A, Arnstein P, Abbey NW, Demartis F, Cohen RL, Shu man MA, Levy JA. (1994). Characterization of a human Kaposi's sarcoma cell line that induces angiogenic tumors in animals. AIDS 8:575-81. Hidalgo, M., and Eckhardt, S. G. (2001). Development of matrix metaliopro 5 teinases inhibitors in cancer therapy. J. NatI. Cancer Inst. 93: 178-193. Hirschmann, R., Yao, W., Arison, B., Maechler, L., Rosegay, A., Spengeler, P.A., and Smith, A.B. (1998), Synthesis of the first tricyclic homodetic peptide. Use of Coordinated Orthogonal Deprotection to Achieve Directed Ring Clo sure. Tetrahedron 54 (1998) 7179-7202. 10 Houghten, R.A., Pinilla C., Appel J.R., Blondelle S.E., Dooley C.T., Eichler J., Nefzi A., Ostresh J.M. Mixture-based synthetic combinatorial libraries. J. Med. Chem. 1999;42:3743-78. Mase K, lijima T, Nakamura N, Takeuchi T, Onizuka M, Mitsui T, Noguchi M. Intrabrochial orthotopic propagation of human lung adenocarcinoma - charac 15 terizations on tumorigenicity, invasion and metastasis. Lung cancer 36 (3): 271-276, 2002. Mergler, M. and Durieux, J.P., BACHEM AG, 2000 by BACHEM AG, CH-4416 Bubendorf, Switzerland. Naknishi, H, and Kahn, M. (1996). Design of peptidomimetics. In: The practice 20 of medical chemistry, pp. 571-590. Academic Press Nargund, R.P., Patchett, A.A., Bach, M.A. Murphy, M.G., Smith, R.G. (1998) Peptidomimetic growth hormone secretagogues. Design considerations and therapeutic potential. J.Med. Chem. 1998; 41:3103-27. Nicklin, S. A., White, S. J., Watkins, S. J., Hawkins, R. E., Baker, A. H. (2000). 25 Selective targeting of gene transfer to vascular endothelial cells by use of pep tides isolated by phage display. Circulation 102: 231-237. Novabiochem # 1 for innovation. Novabiochem 2000 Catalog, Calbiochem Novabiochem AG, Weidenmattweg 4 L~ufelfingen, Switzerland, 2000. Peptide and Peptidomimetic Synthesis, Reagents for Drug Discovery, Fluka 30 ChemieGmbH, Buchs, Switzerland, 2000. Prochiantz, A. (1996). Getting hydrophilic compounds into cells: lessons from homeopeptides. Curr. Op. Neurobiol. 6: 629-634. Promega Notes Magazine, Promega Corporation, Number 74, InCELLectTM Cell-Permeable Peptides, 2000. 35 Protective Groups in Organic Synthesis, Third Edition, Theodora W. Greene, Peter G.M. Wuts, 1999, John Wiley & Sons, Inc. ISBN: 0-471-16019-9.

WO 2004/031218 PCT/F12003/000723 93 The BACHEM Practise of SPPS, (2000) Tips and tricks from the experts at BACHEM, compiled by Mergler, M. and Durieux, J.P., BACHEM AG, 2000 by BACHEM AG, CH-4416 Bubendorf, Switzerland. The Combinatorial Chemistry Catalog & Solid Phase Organic Chemistry 5 (SPOC ) Handbook, Novabiochem # 1 for innovation, Switzerland, Calbio chem-Novebiochem AG Weidenmattweg 4, CH-4448 L~ufellingen, Switzer land, 1998-1999. Welch, D. R., Bisi, J. E., Miller, B. E., Conaway, D., Seftor, E. A., Yohem, K. H. et al. (1991). Characterization of a highly invasive and spontaneously metas 10 tatic human malignant melanoma cell line. Int. J. Cancer 47: 227-237. Yue, C., Thierry, J. and Potier, P. (1993) 2-phenyl isopropyl esters as carboxyl terminus protecting groups in the fast synthesis of peptide fragments, Tetrahe dron Letters 34(2): 323-326.

Claims (26)

1. A tumor targeting unit comprising a peptide sequence: Cy-Rrn-Dd-Ee-Ff-Rrm-Cyy 5 or a pharmaceutically or physiologically acceptable salt thereof, wherein, Dd-Ee-Ff is Aa-Bb-Cc or Cc-Bb-Aa, wherein 10 Aa, is isoleucine, leucine or tert-leucine, or a structural or functional analogue thereof; Bb is arginine, homoarginine or canavanine, or a structural or func tional analogue thereof; Cc is serine or homoserine, or a structural or functional analogue 15 thereof; Rr are each, independently, any amino acid residue or structural or functional analogues thereof; n and m, independently, are integers 0 - 4 and the sum of n and m does not exceed four, and, 20 Cy and Cyy are entities capable of forming a cyclic structure by means of an amide or ester bond.

2. The tumor targeting unit according to claim 1, wherein the peptide is cyclic or forms part of a cyclic structure.

3. The tumor targeting unit according to claim 2, wherein a cyclic 25 structure is formed by a lactam or lactone bond.

4 The tumor targeting unit according to claim 3, wherein one of Cy and Cyy is aspartic acid, glutamic acid or a structural or functional analogue therof, and the other is lysine, ornithine or a structural or functional analogue thereof. 30

5. The tumor targeting unit according to any one of claims I to 4, wherein the sum of n and m is 2 4.

6. The tumor targeting unit according to any one of claims 1 to 4, wherein either n or m or both are zero.

7. The tumor targeting unit according to any one of claims 1 to 6, 35 wherein Rr is any amino acid residue, except histidine or lysine. WO 2004/031218 PCT/F12003/000723 95

8. The tumor targeting unit according to claim 7, wherein Rr is se lected from the group consisting of glycine, arginine and structural or functional analogues thereof.

9. The tumor targeting unit according to any one of claims 1 to 8, 5 wherein Dd-Ee-Ff is LRS or SRL.

10. The tumor targeting unit according to claim 9, having the for mula DLRSK (SEQ ID NO. 1), DGRGLRSK (SEQ ID NO. 2), DRGLRSK (SEQ ID NO. 3), DRYYNLRSK (SEQ ID NO. 4), DSRYNLRSK (SEQ ID NO. 5), DLRSGRK (SEQ ID NO. 6) DLRSGRGK (SEQ ID NO. 7), OLRSE (SEQ ID 10 NO. 8), OLRSGRGE (SEQ ID NO. 9) or KLRSD (SEQ ID NO. 10).

11. The tumor targeting unit according to any one of the previous claims, wherein the unit is derivatized, activated, protected, resin bound or other support bound.

12. A tumor targeting agent comprising at least one targeting unit of 15 any one of claims 1 to 11, coupled to at least one effector unit.

13. The tumor targeting agent according to claim 12, wherein the ef fector unit is a directly or indirectly detectable agent or a therapeutic agent.

14. The tumor targeting agent according to claim 14, wherein the detectable agent comprises an affinity label, a fluorescent or luminescent label, 20 a chelator, a metal complex, an enriched isotope, radioactive material or a paramagnetic substance.

15. The tumor targeting agent according to claim 15, wherein the detectable agent is a rare earth metal.

16. The tumor targeting agent according to claim 13, wherein the 25 therapeutic agent is selected from the group consisting of cytotoxic and cy tostatic substances and radiation emitting substances.

17. The tumor targeting agent according to claim 20, wherein the detectable agent is gadolinium.

18. The tumor targeting agent according to claim 17, wherein the 30 therapeutic agent comprises doxorubicin, daunorubicin, methotrexate or boron.

19. The tumor targeting agent according to any of claims 12 - 18, further comprising an optional unit.

20. A diagnostic or pharmaceutical composition comprising at least one targeting unit according to any one of claims I to 11, or at least one target 35 ing agent according to any one of claims 12 to 19. WO 2004/031218 PCT/F12003/000723 96

21. Use of a targeting unit according to any one of claims I to 11, or a targeting agent according to any one of claim 12 to 19 for the preparation of a medicament for the treatment of cancer or cancer related diseases.

22. The use according to claim 21, wherein said cancer or cancer 5 related disease is a solid tumor.

23. The use according to claim 22, wherein said cancer is selected from the group consisting of carcinoma, sarcoma, melanoma or metastases.

24. A method for treating cancer or cancer related diseases, com prising providing to a patient in need thereof a therapeutically effective amount 10 of a pharmaceutical composition according to claim 20.

25. The method according to claim 24, wherein said cancer or can cer related disease is a solid tumor.

26. The method according to claim 25, wherein said solid tumor is selected from the group consisting of carcinoma, sarcoma, melanoma or me 15 tastases.