Tissue-derived tumor growth inhibitors, methods of preparation and uses thereof

The present invention provides a method for inhibiting cytotoxic poisoning of normal cells in a subject which comprises administering to the subject an amount of a Transforming Growth Factor Beta effective to slow the growth of normal cells and thereby inhibit the cytotoxic poisoning of the normal cells in the subject. Typically, administration is topical and initiated prior to antineoplastic therapy such as radiation treatment or chemotherapy. The invention is particularly suited for pediatric patients.

(Referred to in PCT Gazette No. 12/1994, Section II)

WO 94/06459

PCT/US93/08937

TISSUE-DERIVED ‘ TUMOR GROWTH INHIBITORS, METHODS OF PREPARATIONRND USESTHEREOF

This application is a continuation-in-part· of U.S. Serial No. 992,479, filed December 15, 1992, which is a 10 continuation-in-part of U.S. Serial No. 948,005, filed September 21, 1992. The contents of all of the aboveidentified applications are incorporated in their entireties into the present application.

Throughout this application, various publications are referenced by author and date in parentheses. Full citations for these publications may be found listed alphabetically at the end of the specification immediately preceding the claims. The disclosures of these publications 20 in their entireties are hereby incorporated by reference into this application in order to more fully describe the state of the art as known to those skilled therein as of the date of the invention described and claimed herein.

Background of the invention

Currently, 800,000 patients receive chemotherapy in the United States each year. Myelosuppression and mucositis are common major side effects in conventional chemotherapy and 30 are exacerbated by high dose and/or schedule intensification. As chemotherapeutic dosage has been identified as a major predictor of treatment outcome in a variety of cancer treatment programs, overcoming dose limiting toxicity has become a major goal for oncologists. 35 Thus, to allow intensification of chemotherapy, it will be essential to prevent these unwanted side .effects.

<img file="AU683024B2_D0001.tif" />

Studies have shown that both oral and gastrointestinal toxicity are a major problem in the clinic (Guggenheimer et al., 1977; Lockhart et al., 1981; Sonis et al., 1988; and Sonis &amp; Clark, 1991). This toxicity is very common and has 5 become dose or schedule limiting in both standard and high dose chemotherapy. As improvements are made in management of hematopoietic toxicity, oral toxicity is more frequently dose and schedule limiting. For example, 67% of leukemia/lymphoma patients, 75% of patients receiving bone 10 marrow transplants following therapy, and approximately 20% of breast cancer patients, suffer from oral mucositis. It has been estimated that 40% of chemotherapy patients develop oral complications (Sonis et al., 1978). Many colon cancer patients receiving 5-fluorouracil (5-FU) suffer from oral 15 mucositis.

Chemotherapy agents which cause oral and gastrointestinal mucositis include commonly prescribed agents. Single-agent or combination therapies which result in dose or schedule 20 limiting oral and gastrointestinal (GI) toxicity include, but are not limited to, 5-FU, methotrexate, doxorubicin, Melphalan (see also Table A), and regimens such as the AFM regimen, the CEA regimen and the CAF regimen (Sonis et al., 1990). Cytotoxic agents, such as antiviral agents, also 25 cause mucositis.

Oral mucositis following chemotherapy is a consequence of the high proliferative rate of the normal epithelial cells lining the oral cavity. Chemotherapy-induced mucositis in 30 the oral cavity is largely derived from injury to the basal epithelial cell layer. The epithelial lining of the mouth is generally five cells thick, with the self-renewing stem cell layer located at the base, which itself overlays a fibrous connective tissue matrix. In contrast to skin, most 35 of the epithelium which lines the oral cavity is not highly

<img file="AU683024B2_D0002.tif" />

WO 94/06459

PCT/US93/08937

-3keratinized, and therefore is able to absorb and elicit a biological response to exogenously added peptide growth factors. The epithelium at the base of the mouth divides most rapidly and is most sensitive to chemotherapy.

The extent of oral mucositis appears dependent on the cycling status of the epithelial cell layer. For example, 90% of pediatric chemotherapy patients (ages 1-20) develop oral mucositis, as compared with 18% of patients over the age of sixty (Sonis et al., 1979). Further, in animal models of oral mucositis, increasing proliferation of the oral epithelium by prior administration of most growth factor (e.g. EGF,.TGF-a) markedly increases the severity of oral mucositis. These positively acting factors are currently being evaluated for wound healing.

Oral mucositis leads to epithelial thinning and ulceration resulting in severe pain, weight loss from failure to eat or drink (often requiring parenteral feeding), infection (bacterial, fungal and viral), fever, nausea and diarrhea. Symptoms peak 7 to 10 days following therapy, and gradually recede over the following two weeks. Another majorcomplication is potentially life-threatening infection due to sepsis. Therefore, methods of preventing oral toxicity would enable escalated schedules and doses of a chemotherapy regimen, resulting in improved patient long term survival for many human cancers, notably breast, lung and testicular cancer, leukemia, lymphoma, and neuroblastoma.

Chemotherapy-induced mucositis in the intestinal tract (gastrointestinal mucositis) is also marked and often dose limiting. In the development of the intestInal crypt, primitive stem cells in the proliferative zone at the base of the crypt give rise to more differentiated cell progeny which rapidly migrate to the top of the crypt. These more

WO 94/06459

PCT/US93/08937 ·

-4highly differentiated transitional cells have increasingly less capacity for self renewal and are eventually sloughed off at the top of the crypt. Cell proliferation is restricted to a centrally located band of cells (about ten cell layers deep) within the crypts. Within this band the cells divide rapidly (every ten to fourteen hours) and the product of this division activity matures and moves out of the crypt, onto the villus and is eventually lost within two to three days from the villus tip. Damage to the GI tract increases the risk of sepsis. Thus, like oral mucositis, there is a need to prevent gastrointestinal toxicity.

TGF-Β is part of a. family of multifunctional proteins which modulate, alone or in combination with other molecules, cell proliferation and differentiation. Mature TGF-β is produced from a precursor form, consisting of a pro-region and mature TGF-β. Known isoforms of mature TGF-β are designated TGFBl, TGF-B2, TGF-B3, TGF-B4, and TGF-B5 (Brunner 1988). Mature TGF-β has been isolated from various species (See EP 105,211 and U.S. Patent No. 5,104,977). Across species TGFBs e.g. murine, bovine, human, and porcine show very little difference in amino acid composition. The expression of TGF-β in both normal and transformed cells, and methods of producing biologically active mature TGF-β from eucaryotic cells have been described (Madisen et al., 1988, Pinaher et al., 1585, Assoian et al., 1983, Wrann et al., 1987).

TGF-α and TGF-β are not homologous structures at the protein level and do not exhibit any sequence conservation at the nucleic acid level. As shown in Table 1 below, their physical and biological properties are also different.

WO 94/06459

PCI7US93/0S937

-5TABLE 1

COMPARISON OF PROPERTIES OF TGF-α and TGF-β

Phy,s„igaJ,.. pyppegtieju.

TGF-o

Quaternary structure Subunit (amino acids) monomer

TGF-B dimer 112

Biological Properties:

EGF receptor-binding Mitogenic activity in epithelial cells Arrests cell cycle in epithelial cells

Stimulates fetal bone and collagen production

The ability of growth factors to accelerate healing is well known. Recently, Florine et al. U.S. Patent No. 5,102,870, issued April 7, 1992, described the use of positively acting growth factors and combinations to aid in the healing of oral mucositis. In contrast, TGF-β is a negative regulator of epithelial cell division.

Myelosuppression has been a major dose-limiting toxicity in cancer chemotherapy, and antiviral therapy, e.g. AZT in AIDS therapy or management, which derives from the high proliferative rate of the hematopoietic compartment. The bone marrow contains a totipotent stem cell population that can give rise to a wide variety of cell types involved in oxygen transport, defense against infection. Granulocyte and neutrophil progenitor cells proliferate rapidly and their mature counterparts live only one to two days in the circulation. Consequently, 10-14 days after initiation of chemotherapy, the granulocyte and neutrophil population fall rapidly to a nadir. A reduction in megakaryocyte and erythroid precursor cell populations are also observed but

WO 94/06459

PCT/US93/08937

-6this is less pronounced due to the longer life of their differentiated progeny. Thus, in many of the conventional protocols, chemotherapy is . scheduled to maximize granulrs/te/neutrophil recovery and to minimize infection.

Recent animal model data suggest that the ability to reconstitute the mature blood cell types decreases progressively with multiple rounds of chemotherapy, and administration of either G-CSF or GM-CSF accelerates the loss of self renewal capacity (Hornung and Longo, 1992). This phenomenon appears to result from progressive damage to the hematopoietic stem cell population. Although stem cells are normally quiescent, they are induced into active division following the first round of chemotherapy and thus become highly susceptible to the subsequent administration of cytotoxic drugs. Extensive experimentation has c?^arly demonstrated the phenomenon of stem cell depletion following multiple rounds of chemotherapy in animal models. This loss of hematopoietic renewal was one of the major driving forces behind the development of autologous bone marrow transplantation.

Amento et al. U.S. Patent No. 5,108,989 issued April 28, 1992 discloses the systemic use of TGF-β for predisposition of mammals to accelerated tissue repair. In contrast the present invention involves a method for .inhibiting the cytotoxic poisoning of normal cells by slowing*their growth.

WO 94/06459

PCT/US93/08937

-7summary of the Invention

The present invention provides a method for inhibiting cytotoxic poisoning of normal cells in a subject which 5 comprises administering to the subject an amount of a Transforming Growth Factor Beta effective to slow the growth of normal cells and thereby inhibit the cytotoxic poisoning of the normal cells in the subject. Typically, administration is topical and initiated prior to anti10 neoplastic therapy such as radiation treatment or chemotherapy. The invention is particularly suited for pediatric patients.

WO 94/06459 PCT/US93/08937 ' —8Brieff description off the Picruges

Figure 1: Evaluation of TGF-03 as an chemoprotective agent of epithelial cells in vitro, challenging mink cells with 5 colchicine and vinblastine.

Figure 2 A-C: Hamster cheek pouches on Figure 2(A) day 9;

Figure 2(B) day 11; and Figure 2(C) day 13 post primary treatment with 5-FU, with placebo treatment (Group 1), 2gg 10 TGF-B3 topical treatment (Group 2), or 20Mg TGF-B3 topical treatment (Group 3).

Figure 3: Mean mucositis scores on days 5 through 16 of hamsters treated i.p. with 5-FU on days 0 and 2 at doses of 15 60 mg/kg and 40 mg/kg respectively. Group 1: (Open circles) treatment with 5-FU alone. Group 2: (Open triangles) treatment with 5-FU and 2gg TGF-B3 topically. Group 3; (Open squares) treatment with 5-FU and 20Mg TGF-B3 topically. Group 6 (Filled circles) treatment with 5-FU and 2 0 4/ig TGF-B3 intraperitoneally.

Figure 4: Statistical significance data for Group 2: (Open circles) treated with 5-FU alone. Group 2: (Open triangles) 5-FU with 2/xg TGF-B3 topically. Error bars 25 represent the standard error about the mean mucositis score.

Figure 5: Statistical significance data for Group 3: (open circles) treated with 5-FU alone. Group 3: (open triangles) 5-FU with 20gg TGF-B3 topically. Error bars 30 represent the standard error about the mean mucositis score.

Figure 6: Survival data days 4-16 for Group 1: (open circles) treated with 5-FU alone. Group 2: (open triangles) 5-FU + 2gg TGF-63 topically. Group 3: (open

WO 94/06459

PCI7US93/08937

-9squares) 5-FU and 20gg TGF-B3 topically. Group 6: (solid circles) 5-FU and 4μ$ TGF-B3 intraperitoneally.

Figure 7s Average weight curves for days 4-16. Group Is (open circles) treated with 5-FU alone. Group 2s (open triangles) 5-FU and 2gg TGF-B3 topically. Group 3: (open squares) 5-FU and 2 Ogg TGF-B3 topically. Group 6: (filled circles) 5-FU and 4μς TGF-B3 intraperitoneally.

Figure 8s Mean mucositis scores for Group Is (open circles) 5-FU alone. Group 4: (open triangles) 5-FU and additional 2gg TGF-B3 topically administered post-5-FU, twice daily on days 5, 6 and 7.

Figure 9s Mean mucositis scores for Group 1 (open circles): 5-FU alone. Group 5: (open triangles) 5-FU and additional 20gg TGF-B3 topically administered post 5FU, twice daily on days 5, 6 and 7

Figure 10-11s Effects of TGF-B3 on mouse gastrointestinal epithelium in vivo. Figure 10A-10D shows BDF mice four days following second 5-FU injection with and without TGF-B3. Figure 11A-11F shows Balb/c mice four days after second 5-FU injection.

Figure 12A-12C: Figure 12A and 12C show enhanced survival of 5-FU treated mice by TGF-B3 with cytokines G-CSF and IL-1 (-·- 5-FU alone; -&lt;&gt;- 5-FU and cytokines; -x- 5-FU and cytokines and TGF-B3). Figure 12B shows enhanced weight retention of TGF-B3 treated mice undergoing 5-FU chemotherapy (-®- control; —_B— cytokines; -ο- 3X TGF-B3 2 gg with cytokines; -_A- 4X TGF-B3 2 μ$ with cytokines; -_π- 6X TGF-B3 2 gg with cytokines).

SUBSTITUTE SHEET (RULE 26)

WO 94/06459

PCT/US93/08937

I

-10Figure 13: Inhibition of hematopoietic stem cells by Transforming Growth Factor Beta. Primary bone marrow cells were enriched for a progenitor stem cell population by immunodepletion and cultured as described. Cells were grown in the presence of Mo-T Cell conditioned media and increasing concentrations of purified TGF-/J3 (0, 10, 100, 1000 pM) and colony sizes determined at 3, 7 and 14 days. TGF-03 inhibited proliferation and subsequent increase in hematopoietic colony size in a dose and time dependent manner.

Figure 14: . Figure 14 shows a theoretical plot of TGF-B3 chemoprotection from high dose chemotherapy (--- chemotherapy only; ---- chemotherapy + TGF-B3).

Figure 15: Demonstration of in vitro time dependence of chemoprotection by TGF-B3 of early hematopoietic stem cells (HPP-CFU) from Ara-C.

Figure 16: Demonstration of in vitro concentration dependent chemoprotection by TGF-B3 of early hematopoietic stem cells (HPP-CFU) from Ara-C.

WO 94/06459

PCT/US93/08937

-11Detailed Description of the invention

As used herein, TGF-β means a protein which exists in vivo as a covalently linked dimer having a molecular weight of approximately 25,000. The term TGF-β means and includes natural and recombinant TGF-Bl (TIF, TIF-1),. TGF-B2, TGF-B3 (TGI, TGI—1, TGI-2),' TGF-B4, TGF-B5, or TGF-β heterodimers including TGF-Bl.2 and TGF-B2.3. The natural TGF-B may be purified from tissue such as bone, platelets, or placenta or, more' preferably prepared using recombinant technology (Iwata et al. 1992 A; ten Dijke et al. 1990 A; ten Dijke et al. 1990 B; Purchio et al., UK Patent Publication No. GB2210620A; Detynck et al., U.S. Patent No. 4,886,747; Jakolaw et al., 1988; Cerletti, N., 1991; Ogawa et al., 1992, McMaster et al. EP 542 679).

As used herein, Transforming Growth Factor Beta means and includes 1) TGF-B; 2) a biologically active homolog of TGFB; 3) biologically active mutants of TGF-B; and 4) biologically active variants of TGF-β; and 5) biologically active fragments of any of 1) to 4). In this respect biologically active means having substantially the same biological activity as native TGF-β, e.g. binding to Type I or Type II TGF-B receptors and inhibiting Mink Lung cell line CCL64. Mutants and variants may be polypeptides in which the amino acid sequence has been changed but are still biologically active including deletion mutants, substitution mutants, variants with amino acid substitutions. In addition, Transforming Growth Factor Beta means and includes variants capable' of binding to the Type I or Type II receptors in which the biological activity has been mediated. Also, conjugates of Transforming Growth Factor which have altered biological properties such as increased half-life in vivo, increased activity, or both.

WO 94/06459

PCT/US93/08937

-12As used herein, stem cells include epithelial stem cells and hematopoietic stem cells. Hematopoietic progenitor cells are pluripotent cells capable of self renewal or differentiation into multiple hematopoietic cell lineages 5 (e.g. erythroid cells, myeloid cells, etc.).

the present invention provides a method for inhibiting cytotoxic poisoning of normal cells in a subject which comprises administering to the subject an amount of a 10 Transforming Growth Factor Beta effective to slow the growth of normal cells and thereby inhibit the cytotoxic poisoning of the normal cells in the subject. The normal cells may be epithelial cells, oral’ epithelial cells, epithelial cells lining the esophageal tract, lung epithelial cells, 15 capillary endothelial cells, type I lung epithelial cells, gastrointestinal epithelial cells, colon crypt stem cells, stem cells, hematopoietic progenitor cells, or enterocytic cells (see Sonis, 1990, Stover, 1990).

Cytotoxic poisoning may be associated with antineoplastic therilpy such as chemotherapy, or radiation therapy. Alternatively, cytotoxic poisoning may be a result of antiviral, anti-bacterial, anti-fungal, anti-microbial, antiparasitic, or other therapeutics. The cytotoxic poisoning may be due to antiviral agents such as AZT, which causes mucositis (Posner, 1990).

In the present invention, the administration is initiated preferably prior to anti-neoplastic therapy fcr the subject.

The anti-neoplastic therapy may be either chemotherapy or radiation therapy. Alternatively, the therapy may be any cytotoxic therapy. The invention is particularly suited to pediatric subjects. The administration may continue through a cycle of chemotherapy, for up to five days in order to 35 retain its protective effect.

WO 94/06459

PCT/US93/08937

-13Radiation therapy is widely used in the treatment of various tumors but has many side effects including loss of rapidly proliferating epithelial cells e.g. of the oral cavity, GI tract, and the scalp. This leads to severe mucositis and alopecia in many cases.

Transforming Growth ’ Factor Beta is useful in preventing epithelial cell damage during chemotherapy. Transforming Growth Factor Beta inhibits the proliferation of a variety of normal epithelial cell types and hematopoietic stem cells, in contrast to tumor cells which are often impaired in their ability to transduce signals mediated by TGF-B binding to type I or type TGF-B receptors. Thus, tumor cells are generally less responsive to the growth inhibiting effects of TGF-B3. The ability of TGF-B3 to reversibly inhibit cycling of normal hematopoietic stem cell or epithelial cell types, but not many tumor cell types, allow TGF-B to be used to as a chemoprotectant· of normal tissue during cytotoxic chemotherapy. TGF-B3 mediated protection of the mucosa is valuable in reducing severe and painful mucositis. Implementation of this therapeutic approach represent a major advance in cancer management.

Growth modulation of hematopoietic cells by Transforming Growth Factor Beta is highly specific. Early hematopoietic stem cells stimulated by IL-1 and kit ligand (stem cell factor SCF, mast cell growth factor, MGF) are inhibited by Transforming Growth Factor Beta, whereas the proliferation of later cell lineages stimulated by GM-CSF are potentiated by Transforming Growth Factor Beta. Based on in vitro and in vivo experiments, Transforming Growth Factor Beta plays a pivotal role in the ultimate goal of coupling stem cell expansion and protection strategies with accelerated effector cell regeneration.

I':\OI&gt;liR\MRO\5l&lt;i28.93.SI’E-18/7/97

-14• « · ¹ ·« •« •· β ··· • · • »

Transforming Growth Factor Beta may be administered in. several ways to effect hematopoietic chemoprotection. Transforming Growth Factor Beta is administered during or immediately prior to chemotherapy to protect an early hematopoietic stem cell population or normal epithelial cell population in multicycle or accelerated cycle chemotherapy. Hematopoietic growth factors may be used in the recovery phase to accelerate repopulation of the macrophage/granulocyte lineages. More particularly, a mitogen for epithelial cells, such as but not limited to EGF, TGF-α, amphiregulin, FGF. IGF-1, or other mitogen for epithelial cells may be contacted with the TGF-P-treated epithelial cells following antineoplastic therapy. Alternatively, or in addition, a cytokine such as G-CSF, GM-CSF, MCSF, IL-3, IL-1, IL-6 or kit ligand may be used to accelerate repopulation during the recovery phase. Transforming Growth Factor Beta treatment prevents long term stem cell depletion, neutropenia and produces a continued patient tolerance to further cycles of chemotherapy.

• · • · β e • · ·· • ·· » ·· β · • · ··

In addition, hematopoietic growth factors can be administered prior to chemotherapy to effect an increase in total bone marrow cellularity and effect an increase the number of hematopoietic cells surviving cytotoxic chemotherapy. However, expanding the hematopoietic stem cell population using positive growth factors (for example kit ligand or stem cell factor) requires a mechanism to restore these cycling cells back to a non-dividing state prior to chemotherapy. Transforming Growth Factor Beta administered to this expanded population of hematopoietic stem cells prior to chemotherapy acts to protect and increase the reserve of the more mature hematopoietic cells. Thus, Transforming Growth Factor Beta may reduce the fall in white blood cell count associated with chemotherapy.

For oral mucositis, the preferably administration is topical. Dose ranges for topical use include 1 ng to 10 mg per dose, preferably, 100 ng to 2 mg per dose or more preferably 10 /zg to 500 μ§, per dose. However, the administration may also be systemic. Dose ranges for systemic application include 1 ng to 500 /zg/kg subject weight/day, preferably 80 ng to 100 /zg/kg subject

<img file="AU683024B2_D0003.tif" />

WO 94/06459

PCT/US93/08937

-15weight/day, or mote preferably ι μς to 20 gg/kg subject weight/day.

Chemotherapy agents which cause mucositis include, but .are not limited to 5~FU, Ara-C, doxorubicin (Adriamycin) daunorubicin, methotrexate, vincristine, vinblastine, Melphalan, cytosine arabinoside, thioguanine, bleomycin, dactinomycin, cisplatin, mithramycin, mitomvcin, hydroxyurea and procarbazine hydrochloride (See also Table A) . Patients receiving either high dose or accelerated cycle chemotherapy are at risk of developing mucositis, as are pediatric patients, patients undergoing bone marrow transplantation and patients receiving radiation therapy, especially those receiving therapy for squamous carcinoma of the head and neck.

Alternatively, administration may be by tissue-specific, locally activated mutant proTGF-Bs. For example, the mutant TGF-B3 precursor is expressed as a single homodimeric polypeptide in a host cell by mutation of the R-K-K-R cleavage site between the TGF-B3 pro region and mature TGF-B3 to a protease cleavage site, e.g. factor Xa cleavage sequence (Ile-Glu-Gly-Arg) or a collagenase cleavage sequence (Pro-X-Gly-Pro) using standard site directed mutagenesis procedures, followed by insertion of the mutant TGF-B3 nucleic acid into a expression vector and transfection of the mutant TGF-B3/vector DNA into a host cell (e.g. E. coli, CHO, or HeLa) together with DNA encoding a selectible marker (e.g. neo, dhfr). Alternatively, the cleavage site may be any specifically cleaved sequence as exemplified but not limited to the following examples: Ala64-Subtilisin (His-X-X-X-Trp) , a chymotrypsin (Tyr-Phe), a enteropeptidase (Gly[Asp]j_₅ Lys), a Factor Xlla (Ile-ValGly-Gly-Thr-Val), a plasma kallikrein (Pro-Phe-Arg), or a thrombin (Gly-Pro-Arg) (Nilsson et al., 1988). These

WO 94/06459

PCT/US93/08937 '

-16 — locally activated Mutant proTGF-Bs may be used to deliver TGF-B to a precise organ target. The mutant proTGF-Bs may reduce toxicity associated with non-specific TGF-β dose regimens. Similar mutants may be made for TGF-Bl, TGF-;B2, TGF-B4, or TGF-B5.

An effective amount as used herein refers to that amount which provides a prophylactic effect by preventing cytotoxic poisoning of normal cells or mucositis in a subject receiving a given condition and administration regimen. Pharmaceutically-acceptable acid addition salts of the compounds of this invention include, but are limited to, those formed with HC1, HBr, HNO,, H₂SO₄, H₃PO₄, CH₃SO₃H, p~ CH₃C₆H₄SO₃H, CH₃CO₂H, gluconic acid, tartaric acid, maleic acid and succinic acid. Pharmaceutically-acceptable cationic seilts of the compounds of this invention include, but are not limited to, those of sodium, potassium, calcium, magnesium, ammonium, N,Ν'-dibenzylethylenediamine, Nmethylglucamine (meglumine), ethanolamine and diethanolamine.

For administration to humans in either a curative or prophylactic treatment of mucositis or other conditions, oral dosages of the compounds are generally in the range of from 0.1-100 mg daily for an average adult patient (70 kg). Thus, for a typical adult patient, individual tablets or capsules contain from 0.1 to 50 mg of active compound, in a suitable pharmaceutically acceptable vehicle or carrier. Dosages for intravenous administration are typically within the range of 0.1 to 10 mg per single dose as required. For intranasal or inhaler administration, the dosage is generally formulated as a 0.1 to 1% (w/v) solution. In practice, the physician will determine the actual dosage which will be most suitable for an individual patient and it will vary with the age, weight, and response of the

WO 94/06459

PCT/US93/08937

-17particular patient. The above dosages are exemplary of the average case but there can, of course, be individual instances where higher or lower dosage ranges are merited, and all such dosages are within the scope of this invention.

For human use, the compounds of this invention can be administered alone, but will generally be administered in an admixture with a pharmaceutical diluent or carrier selected with regard to the intended route of administration and standard pharmaceutical practice. For example, they may be administered orally in the form of tablets containing such excipients as starch or lactose, or in capsules or ovales either alone or in an admixture with excipients, or in the form of elixirs or suspensions containing flavoring or coloring agents. They may be^- injected parenterally; for example, intravenously, intramuscularly or subcutaneously. For ’ parenteral administration, they are best used in the form of a sterile aqueous solution which may contain other substances; for example, enough salts or glucose to make the solution isotonic.

In addition, compositions comprising prophylactically effective amounts together with suitable diluents, preservatives, solubilizers, emulsifiers, adjuvants and/or carriers useful for Transforming Growth Factor Beta therapy. Such compositions are liquids or lyophilized or otherwise dried formulations and include diluents of various buffer content (e.g., Tris-HCl, acetate phosphate), pH and ionic strength, additives such as albumin or gelatin to prevent absorption to surfaces, detergents (e.g., Tween 20, Tween 80, Pluronic F68, Pluronic F127, bile acid salts), solubilizing agents (e.g., Thimerosal, benzyl alcohol), bulking substances or tonicity modifiers (e.g., lactose, mannitol), covalent attachment of polymers such as polyethylene glycol to the Transforming Growth Factor Beta,

WO 94/06459

PCT/US93/08937 ·

-18compJexation with metal ions, or incorporation of the material into or onto particulate preparations of polymeric compounds such as polylactic acid, polyglycolic acid, polyvinyl pyrrolidone, etc. . or into liposomes, microemulsions, micelles, unilamellar or multilamellar vesicles, erythrocyte ghosts, or spheroplasts. Such compositions will influence the physical state, solubility, stability, rate of in vivo release, and rate of in vivo clearance of Transforming Growth Factor Beta. Other ingredients optionally may be added such as antioxidants, e.g., ascorbic acid, L-cysteine, sodium sulfite; low molecular weight (less than about ten residues) polypeptides, e.g., polyarginine or tripeptides; proteins, such as serum albumin, gelatin, or immunoglobulins; amino acids; such as glycine, glutamic acid, aspartic acid, or arginine; chelating agents such as EDTA; and sugar alcohols such as mannitol or sorbitol. The choice of composition will depend on the physical and chemical properties of the Transforming Growth Factor Beta. Controlled or sustained release compositions include formulation of lipophilic depots (e.g., fatty acids, waxes, oils). See also U.S. Patent No. 5,213,811 for sustained release formulations. The invention also includes particulate compositions coated with polymers (e.g., poloxamers or. poloxamines). The Transforming Growth Factor Beta may be coupled to antibodies, ligands, antigens, or targeted against tissuespecific receptors. Other embodiments of the compositions of the invention incorporate particulate forms protective coatings, protease inhibitors, or permeation enhancers for various routes of administration, including parenteral, pulmonary, nasal, buccal and oral.

Transforming Growth Factor Beta may be administered by pulmonary absorption using commercially available nebulizers for liquid formulations. The nebulizers may be jet

WO 94/06459

PCT/US93/08937

-19nebulizers or ultrasonic nebulizers. Liquid formulations may be directly nebulized or lyophilized powder nebulized after reconstitution. A liquid formulation may be directly instilled into the nasotracheal or endotracheal tubes in intubated patients. Alternatively, Transforming Growth Factor Beta may be aerosolized using a fluorocarbon formulation and a metered dose inhaler. It may also be inhaled as a lyophilized and milled powder.

As described above, the Transforming Growth Factor Beta may be modified by conjugation with polyethylene glycol. In a typical preparation methoxypolyethylene glycol-succinimydyl succinate (.18.1 mg- 3.63 gmol; SS-MPEG = Sigma Chemical Co. No. M3152), molecular weight approximately 5,000) is dissolved in 0.32 mL deionized water is added to 13.0 mg (1.16μηο1) TGF-B in 1.0 mL 138 μΜ sodium nhosphate, 62 mM NaCl, 0.62 mM sodium acetate, pH 8.0. The resulting solution is shaken gently (100 rpm) at room- temperature for 3 0 minutes. A 1.0 mL aliquot of the reaction mixture (10 mg protein) is applied to Pharmacia Superdex 75 gel filtration column (1.6 X 50 cm) and eluted with 100 mM sodium phosphate, pH 6.9 at a rate of 0.25 mL/min. Fractions are collected and monitored by UV. See also U.S. Patent Nos. 5,206,344, 4,261,973, 4,609,546, and 4,179,337.

Suitable topical formulations include gels, creams, solutions, emulsions, carbohydrate polymers, biodegradable matrices thereof; vapors, mists, aerosols, or other inhalants. Vapors, mists, aerosols, or other inhalants are particularly suited for inhibiting pulmonary toxicity. Transforming Growth Factor Beta may be encapsulated in a wafer, wax, film or solid carrier, including chewing gums. Permeation enhancers to aid in transport to movement across the epithelial layer are also known in the art and include, but are not limited to, dimethyl sulfoxide, sodium dimethyl

WO 94/06459

PCT/US93/08937 ¹

I

-20glycolate, glycols, and sodium glycolate. Suitable vehicles are described above. The Transforming Growth Factor Beta so administered may be immobilized so as to prevent dilution and washing by saliva. Such agents are also known in the art and include but are not limited to methyl-dimethyl sulfoxide, glycols and propylcellulose and the like. The Transforming Growth Factor Beta may also be administered in the form of an oral wash (see Eisen, D, et al., 1990). Alternatively, the Transforming Growth Factor Beta may be administered in a pectin-based gel such as Orabase ™.

Rectal preparations for GI protection are also prepared by conventional techniques, for example by employing an oleaginous base such as a higher fatty acid glyceride e.g., cocoa butter, Witepsols (a semisynthetic base, Dynamite Nobel, Federal Republic of Germany); a medium fatty acid glyceride e.g., Miglyols (Dynamite Nobel); or a vegetable oil e.g., sesame oil, soybean oil, corn oil., cottonseed oil, olive oil. Suitable enteric coating formulations are exemplified by U.S. Patent Nos. 5,104,430 (Dupont-Merck), 5,096,717 (Ciba-Geigy) , 4,713,248 (Benzon) and 5,178,866 (Alza).

The disclosures of these patent specifications are hereby incorporated in their entireties. Further examples of enteric coating formulations are found in European Patent applications EP 475536 ' (Takeda Chemical) and EP 244,380 (Haessle).

For the prevention of alopecia, Transforming Growth Factor Beta will be typically administered systemically, however, it may also be administered topically in a solution, cream, ointment, gel, lotion, shampoo or aerosol adapted for application to the skin. These pharmaceutical compositions will contain about 0.01 μς to about 100 Mg /day/cm²,

WO 94/06459

PCT/US93/08937

-21preferably about 0.1 gg to about 10 gg/day/cm² of active compound in a pharmaceutical carrier.

For topical application various other pharmaceutical formulations or carriers are useful for the administration of the active compound of this invention. Such formulations include, but are not limited to the following: ointments such as hydrophilic petrolatum or polyethylene glycol ointment; pastes which may contain, gums such as xanthan gum; solutions such as alcoholic, aqueous or buffered solutions; gels such as aluminum hydroxide or. sodium alginate gels; albumins such as human or animal albumins; collagens such a human or animal collagens; celluloses such as alkyl celluloses, hydroxyalkyl celluloses and alkylhydroxyalkyl celluloses, for example methylcellulose, hydroxyethyl cellulose, carboxymethyl cellulose, hydroxypropyl methylcellulose, and hydroxypropyl cellulose; polymers such as Pluronic R Polyols exemplified by Pluronic F-127; tetronics such as tetronic 1508; and alginates such as sodium alginate.

The present invention also provides a method of obtaining bone marrow substantially free of actively dividing tumor cells which comprises: (a) contacting bone marrow containing normal hematopoietic cells and actively growing tumor cells with an effective amount of a Transforming Growth Factor Beta such that the growth of the normal hematopoietic cells is temporarily inhibited; (b) subsequently contacting bone marrow with a tumor inhibiting drug under such conditions to permanently prevent growth of tumors cells; and (d) culturing bone marrow so as to permit growth of normal hematopoietic cells thereby obtaining bone marrow substantially free of actively dividing tumor cells.

WO 94/06459

PCT/US93/08937

-22Additionally, the present invention provides a method of obtaining bone marrow substantially free of actively dividing tumor cells. The method comprises (a) contacting bone marrow containing normal hematopoietic cells and actively growing tumor cells with an effective amount of a Transforming Growth Factor Beta; (b) culturing bone marrow of step (a) in the presence of Transforming Growth Factor Beta for a suitable period, under conditions such that terminal cell differentiation and. clearance of tumor cells is permitted; and (e) obtaining bone marrow substantially free of actively growing tumor cells.

This invention is illustrated in the Experimental Detail sections which follow which show that Transforming Growth Factor Beta inhibits cytotoxic poisoning of normal cells in a subject. These sections are set forth to aid in an understanding of the invention but are not intended to, and should not be construed to limit in any way the invention as set forth in the claims which follow.

WO 94/06459

PCT/US93/08937

-23ExmiMEwaiuPETAnts

Example 1: Assays of Transforming Growth Factor Beta Activity In Vitro on Several Human Epithelial Tumor Cell Lines and Human Fibroblasts

Cell growth was determined using the monolayer assay for Transforming Growth Factor Beta described by Iwata (Iwata et al. 1992 B). Non-leukemic cells were subcultured on 96-well tissue culture plates in 100μ1 of media at a seeding density of 2xl0³'cells per well. Cells were maintained and assayed in Dulbecco's modified Eagle's medium containing 10% fetal bovine and 2% L-glutamine. These cells were treated with 25 ng/ml (=1 nM) of TGF-03, pulsed 24 hours with lgCi/ml 5[¹²⁵I]-iodo-2'deoxyuridine when cells in the untreated control wells were 90% confluent and harvested. Table 2 shows inhibition of some human epithelial tumor lines by Transforming Growth Factor Beta. In contrast, normal epithelial cells are strongly inhibited at picomolar concentrations of TGF-83.

TABLE 2

Effects of TGF-/73 (1 nN) on the Growth of Human Epithelial Derived Cancer Cell Lines in Culture

CELL LINE %.....INHIBITION

Human Epithelial Derived Cancer Lines ·

<td> A549</td><td> (lung adenocarcinoma)</td><td> 46</td>

<td> A375</td><td> (melanoma)</td><td> 47</td>

<td> A2058</td><td> (melanoma)</td><td> 88</td>

<td> WiDR</td><td> (colon adenocarcinoma)</td><td> 24</td>

<td> MCF 7</td><td> (breast carcinoma)</td><td> 57</td>

Normal Human

Huf (foreskin fibroblasts) 6

WO 94/06459

PCT/US93/08937 -24Fibroblasts behave in a manner opposite normal epithelial cells in this assay.

In a similar manner, the effect of TGF-Bl, TGF-B2, TGF-B4, and TGF-B5 on human tumor lines is also assayed.

Example 2; In Vitro Protection of Epithelial Cells Against Chemotherapeutic Drugs (Figure 1)

In order to evaluate Transforming Growth Factor Beta as a protective agent in vitro the following experiments were performed.

Mink cells were seeded in 96-well plates at 10³ cells/well in 100μ1 of DMEM supplemented with 10% fetal bovine serum. Wells containing treated cells received 25μ1 of TGF-/33 (50 ng/ml). After 24 hours incubation with TGF-p3, 25μ1 of either colchicine or vinblastine were added. After another 24 hours, the media was removed and the cells washed once with Dulbecco's PBS and fresh complete media added. The cells were incubated for another 7 days.

Cell growth was quantitated by uptake of 5-[¹²⁵I] iodo2'deoxyuridine (¹²⁵IUdR) indicating the amount of ©ell growth as described (Iwata et al., 1992 B) . . As seen in Figure 1, cells preincubated with TGF-/J3. prior to incubation with various- doses of chemotherapeutic drugs (e.g. vinblastine and colchicine) showed significantly more uptake of ¹²⁵IUdR relative to cells which were incubated with the chemotherapeutic drugs without TGF-03. Therefore, cells preincubated with TGF-/J3 were protected from the toxic effects of the chemotherapeutic drugs. Similar results were observed when adriamycin was used as the chemotherapeutic drug. This protective effect would also be expected with

WO 94/06459

PCT/US93/08937

-25any other chemotherapeutic drugs (which preferentially kill actively dividing cells).

For Transforming Growth Factor . Beta to be effective, in protecting normal epithelial cells from chemotherapy it is apparent that the growth of the tumor is preferably less inhibited than that of normal tissues. This is achieved either because tumors are either innately resistant or less sensitive to the growth inhibitory effects of Transforming Growth Factor Beta. Alternatively, the pharmacokinetics of Transforming Growth Factor Beta administration allow a differential effect between normal and tumor cells in vivo.

In a similar manner, TGF-Bl, TGF-B2, TGF-B4, and TGF-B5 is assayed for protective activity.

Example 3: In Vivo Inhibition of oral Epithelial Proliferation by Topical Application of Transforming Growth Factor Beta

We show topical administration of Transforming Growth Factor Beta to the buccal cavity of hamsters acts to decrease the fraction of oral epithelial cells in S-phase, and thereby inhibit cell proliferation. In this example the buccal pouch of Golden Syrian hamsters were treated with Ogg, 2 pg, 20pg or 200gg of TGF-B3 in a saline diluent and subsequently covered with methyl-propylcellulose.

Three topical TGF-B3 applications were made over an eight hour time period. Only the left cheek was treated with TGFB3; the right cheek served as a control. Biopsies of the cheek pouch were obtained approximately 16 hours after the final TGF-B3 treatment. The biopsied oral tissue was formalin fixed, embedded in paraffin, sectioned and immunohistologically stained, with a monoclonal antibody to

WO 94/06459

PCT/US93/08937 · t

-26PCNA (a marker of S-phase cells). Alternatively, Ki67 antibody staining of sectioned tissue or ³H-thymidine labelling of the oral epithelium in vivo may be used to measure cell division in vivo. Sixteen hours post TGF-B3 treatment the following data was obtained (Table 3):

TASLE-l

In vivo Inhibition of Oral Epithelial Proliferation shown by PCNA Staining of Hamster Oral Mucosa after Topical Administration of TGF-B3.

Topical Treatment Mean Inhibition vs. Control

2gg TGF-B3

2Ogg TGF-B3

200gg TGF-B3

22%

42%

46%

These data show Transforming Growth Factor Beta to be a potent inhibitor of the cell cycling of oral epithelium in vivo.

The in vivo activity of TGF-Bl, TGF-B2, TGF-B4, and TGF-B5 is evaluated in a similar manner.

Example 4: Transforming Growth Factor Beta as an Agent Useful for the Prevention of Epithelial Cell Damage Caused by Chemotherapy or Radiotherapy

An animal model is used to evaluate in vivo efficacy of Transforming Growth Factor Beta as an agent useful for prevention of oral mucositis in hamsters induced by the chemotherapeutic agent 5-FU (Sonis st al., 1990). A number of 5-FU schedules may be useful to produce mucositis and assess therapeutic standards. In one protocol, groups of 12, eight week old Golden Syrian hamsters are given 5-FU (60 mg/kg) on days 0, 5 and 10. Alternatively, 5-FU may to administered on days 0 and on day 2 by the intraperitoneal (i.p.) route in doses of 60 mg/kg and 40 mg/kg,

WO 94/06459

PCT/US93/08937

-27respectively. Animals are observed and weighed daily. The animals are anesthetized with ether and the b‘ -al cavity everted and photographed. Photographs of the cheek pouch are numbered, randomized, independently scored blind using a common set of grading photographs. This approach, grading oral mucositis on scale from 1 to 10, allows statistically significant data. In hamsters as in man, oral mucositis peaks around 7 days from initial exposure to 5-FU. Importantly, oral mucositis is worse when the epithelium is actively dividing (e.g. when TGF-α pretreated), mimicking the situation with pediatric patients undergoing chemotherapy.

Prevention of oral mucositis is shown by pretreatment of hamsters with Transforming Growth Factor Beta prior to 5-FU treatment. Mucositis is graded on a scale of 1 to 10 as described above. Schedule and dose relationship to efficacy is demonstrated using the same animal protocol.

In this example TGF-B3 was administered to the hamster oral cavity to prevent mucositis. Transforming Growth Factorbeta was applied either in solution or as a suspended precipitate. For example, a precipitate in a suspension was prepared by addition of 20 pg of TGF-B3 in 2 mM HC1, 0.1% Tween 80 in a volume of 100 pl to 1 ml of phosphate buffered saline (PBS) pH 7.4. The following groups of 10 Golden Syrian hamsters (age approximately 6 week) were evaluated for the severity and duration of mucositis following 5-FU treatment using the days 0 and 2 schedule described above. The following groups were evaluated:

WO 94/06459

PCT/US93/08937 &lt;

-28TABLE 4A

<td> srauB (N = 10)</td><td> TGF-B3</td><td> Dose Schedule</td><td> Route</td>

<td> 1</td><td> 4 X vehicle alone</td><td> over 24 hrs. day -1 to day 0</td><td> topical</td>

<td> 2</td><td> 4 X 2μς</td><td> over 24 hrs. day -I to day 0</td><td> topical</td>

<td> 3</td><td> 4 X 20Mg</td><td> oyer 24 hrs. day -1 to day 0</td><td> topical</td>

<td> 6</td><td> 3 X 4Mg</td><td> over 8 hrs. day 2</td><td> intraperitoneal</td>

These data were statistically significant and conclusively show TGF-B reduces the severity and duration of mucositis in vivo (Figures 2-5). Increased survival (Figure 6) and weight retention (Figure 7) were also observed. This effect was achieved through cell cycle inhibition of the oral epithelium prior to chemotherapy, and not apparently by accelerated wound healing, as shown by Example 5 below.

Groups of ten three to four week old Golden Syrian hamsters were injected i.p. with 80mg/kg 5-FU on day zero and 60mg/kg

5-FU on day 2. TGF-Bl, TGF/32 or TGF/93 was administered to the hamster oral cavity to prevent mucositis. The following groups were evaluated for the severity and duration of mucositis following 5-FU treatment:

' WO 94/06459

PCT/US93/08937

<td colspan="2"> ΙΔΒΕ.Ε...4.β</td><td rowspan="2"> Dose Schedule</td>

<td> Group</td><td> TGF-B1.2 or3</td>

<td> 1</td><td> 4 X vehicle alone</td><td> over 24 hrs. day -1 to day 0</td>

<td> 2 ·</td><td> 4 X 20gg TGF01</td><td> over 24 hrs. day -1 to day 0</td>

<td> 3</td><td> 4 X 2Ogg TGF02</td><td> over 24 hrs. day -I to day 0</td>

<td> 4</td><td> 4 X 20Mg TGF/J3</td><td> over 24 hrs. day -1 to day 0</td>

Route topical topical topical topical

These initial data show that Transforming Growth Factor Be.ta increased survival by day 8. Mucostis score was also decreased. .

ΙΔΕΕΕ-ΑΩ

HeaiL-Survival

42%

70%

70%

60%

The effective amount of Transforming Growth Factor Beta in the carrier vehicle ranges from 0.01 pg to 10 mg depending on the extent or severity of the oral mucositis. Transforming Growth Factor Beta is applied to the buccal cavity in a suitable physiological carrier up to 5 days prior to treatment with one or more cytotoxic chemotherapy agents.

Example 5: Effect of Additional Transforming Growth Factor Beta Administration After 5-FU Treatment

WO 94/06459

PCT/US93/08937 &lt;

-30To evaluate the effects of Transforming Growth Factor Beta post second 5-FU treatment as used for normal wound healing agents, two additional groups were treated in the same way as Groups 2 and 3, but also receiving topical, oral Transforming Growth Factcr Beta following the second dose of 5-FU as shown in Table 5.

TAELE-S

<td> TGF-B3 post 5-FU</td><td> Time of TGF-B3</td><td> Route</td>

<td> 6 X 2qg</td><td> twice daily days 5, 6 and 7</td><td> topical</td>

<td> 6 X 20^g</td><td> days 5, 6 and 7</td><td> topical</td>

Similar experiments are performed with TGF-Bl, TGF-B2, TGFB4, and TGF-B5.

TGF-0 post 5-FU administration was not effective in preventing mucositis as shown in Figures 8 and 9. No significant differences in mucositis scores were observed between groups 1, 4 and 5. Thus, unlike TGF-α, EGF, or PDGF, administration of topical Transforming Growth Factor Beta after chemotherapy has no beneficial therapeutic effect, unlike TGF-B pre-treatment, and therefore is contraindicated. In this setting and dose TGF-B is a negative regulator of epithelial cell growth, not a positive growth factor for oral tissue in contrast to TGF-α, EGF, and PDGF.

Example 6¾ Topical Application of Transforming Growth Factor Beta and Cell Cycling of Hamster Oral Epithelia in vivo

The effect of TGF-B3 on proliferation of oral epithelium was evaluated, both in vitro by measurement of DNA synthesis in

WO 94/06459

PCT/US93/08937 —31 — cultured hamster oral epithelium and in vivo by measurement of bromodeoxyuridine incorporation and PCNA expression in the hamster cheek pouch. First, short term cultures of oral epithelium (&gt;90%) were established in KBM media (Clonetics), 2.5% FCS and incubated with 0, 1, 100 or 1000 pM TGF-B3 for 48 hours. DNA synthesis was measured by addition of ¹²⁵Iodo-deoxyuridine (1 MCi/ml) for an additional 48 hours as previously described (Iwata et al. 1985) . Exposure to 100 pM or 1 nM TGF-B3 reduced label incorporation to 42% and 23% of the untreated control value respectively.

In vivo, the effect of Transforming Growth Factor Beta on cell cycling of oral epithelium was investigated by two methods. First, the fraction of oral epithelial cells undergoing DNA synthesis was measured after topical application of TGF-B3 (4 doses of 20 pg in saline over 24 hours) to the hamster cheek pouch (Table 5).-Hamsters were injected with bromo-deoxyuridine (BrdU; 5 mg, ip) at 0, 16 and 54 hours post TGF-B3 treatment. Animals were sacrificed 2 hours post BrdU administration, and cheek pouch sections (4m) stained with anti-BrdU antibody (Amersham). The proliferative, self renewal component in oral epithelium has been shown to be clustered, separated by non-dividing regions (Gibbs and Casarett, 1972; Bickenbach and MacKenzie, 1934) . BrdU labelled and unlabeled cells from random fields (n&gt;25) containing 2:1 labelled cell were counted under the light microscope and an S-phase fraction or labelling index (LI) was calculated. Topical application of TGF-B3 reduced the LI to 44% and 51% of mock treated hamsters at 0 and 16 hours respectively. Second, the number of cheek pouch epithelial cells expressing proliferating cell nuclear antigen (PCNA) was determined. PCNA is apparently tethered to the nucleus during S-phase in a form resistant to acid extraction and has been used as a general marker of DNA synthesis in tissue sections. Tissue samples were collected

WO 94/06459

PCT/US93/08937 ·

-3216 hours after topical application of 20 μg TGF-B3 to the left cheek pouch. The untreated right pouch served as control. Sections were prepared and developed with anti-PCNA monoclonal antibody, anti-mouse Ig and DAB (Oncogene 5 Science). Cells in the epithelial layer staining either positive or negative for PCNA were counted under the light microscope from multiple fields (n=30). Results are shown in Table 6. Application of 20 pg of TGF-B3 resulted in a reduction in the PCNA positive epithelial cell fraction to 10 69% of the untreated left cheek pouch, which was significant by t-test (p=.0001), Duncan's or Tukey's test.

These data are consistent with the hypothesis that topically applied Transforming Growth Factor Beta may act by 15 temporarily arresting or delaying progression of oral epithelium into S-phase. This concept is consistent with previous results in which hematopoietic progenitor cells treated with TGF-B3 accumulated in the G1 phase of the cell cycle (Strife et al., 1990) and were similarly protected 20 from chemotherapeutic agents in vitro (Lemoli et al., 1992) or in vivo. In contrast, most epithelial or hematopoieticderived malignancies are relatively refractory to growth · inhibition by TGF-β and are apparently compromised in the signalling pathways through which TGF-β causes cell cycle 25 arrest in late Gl. The low systemic exposure of topically applied TGF-Β to intact oral mucosa, coupled with the differential responsiveness of normal and tumor tissue to TGF-Β, suggest broad feasibility of such means of chemoprotection in cancer treatment. We do not exclude 30 additional mechanisms by which TGF-β might moderate chemotherapy associated toxicities, such as modulation of receptor or cytokine expression. Similar experiments are performed with TGF-Bl, TGF-B2, TGF-B4, and TGF-B5.

WO 94/06459

PCT/US93/08937

-33TAPfrE,.......$ '· Effect of topical application of Transforming

Growth Factor Beta on cell cycling of hamster oral epithelium in vivo.

<td></td><td> BrdU</td><td colspan="2"> labelling</td><td colspan="2"> PCNA staining</td>

<td> Group</td><td> LI (±SEM)</td><td> % of</td><td> P-value</td><td> % of</td><td> P-value</td>

<td></td><td></td><td> control</td><td></td><td colspan="2"> control</td>

<td> control</td><td> 4.2 (±0.3)</td><td> -</td><td> -</td><td> . -</td><td> -</td>

<td> 0 hrs.</td><td></td><td></td><td></td><td></td><td></td>

<td> post B3</td><td> 1.8 (±0.2)</td><td> 44%</td><td> 0.0008</td><td> -</td><td> -</td>

<td> 16 hrs</td><td></td><td></td><td></td><td></td><td></td>

<td> post S3</td><td> 2.1 (±0.5)</td><td> 51%</td><td> 0.09</td><td> 69%</td><td> 0.0001</td>

<td> 54 hrs</td><td></td><td></td><td></td><td></td><td></td>

<td> post B3</td><td> 3.8 (±0.4)</td><td> 90%</td><td> 0.27</td><td> -</td><td> -</td>

Clinical Evaluation and .Patient Administration

Based on scheduling data and dosage data derived by modelling in the Golden Syrian hamster as described herein, Phase I trials are conducted to evaluate safety. Little or no human toxicity is expected in acute topical administration of Transforming Growth Factor Beta to the oral cavity. Dose escalation studies and chronic exposure studies may be performed. In Phase II, patients are evaluated for a reduction in the frequency and severity of oral mucositis associated with chemotherapy. . In clinical practice, patients are protected from oral mucositis by topical or systemic administration of Transforming Growth Factor Beta in a suitable carrier prior to chemotherapy. Administration of Transforming Growth Factor Beta may be continued for a period after chemotherapy i.e. that period of pharmacologic exposure to ensure protection of the oral tissue from the chemotherapeutic agent. The dose range of Transforming Growth Factor Beta administered to patients is determined first by modelling in the hamster as described herein and then is refined in Phase I/Phase II trials. The

WO 94/06459

PCT/US93/08937 -34amount of Transforming Growth Factor Beta applied to the oral cavity in these range finding studies is determined empirically. The patient is monitored for oral mucositis according to several well established scoring criteria, for example the ECOG (Eastern Cooperative Oncology Group) score for oral mucositis. The dose or schedule of Transforming Growth Factor Beta administered to the cancer patient may be adjusted in subsequent rounds of chemotherapy, according to the extent or duration of oral mucositis in the first round of therapy.

Example 7Ά: in Vivo Growth Inhibition of Gastrointestinal Epithelia

The growth inhibitory effects of Transforming Growth Factor Beta on primary normal and neoplastic colon organ cultures was evaluated with a view to establishing Transforming Growth Factor Beta as a chemoprotective agent to reduce gastrointestinal toxicity in the treatment of patients with chemotherapeutic drugs.

Proliferation of colon cells in various stages of differentiation is measured by the incorporation of either [³H]-thymidine or bromodeoxyuridine (BrdU) into colon biopsy organ cultures followed by sectioning and staining of the intact crypt. A 1 mm biopsy specimen is gently washed in DMEM, 10% FCS, 37°C. Three micron sections are cut to avoid unequal distribution of radiolabel. Incubation with TGF-B3 at 0.1 pM, 10 pM, l nM, 100 nM and 10 μΜ is carried out over; several time points.

Colon specimens are labeled in 2 ml of DMEM + 10%, 37°C equilibrated in a 5% CO₂ incubator, containing 1 pCi/Til [3H]-thymidine (20 μσΐ/ππηοίε) or 400 μΜ BrdU for 1 hour. Sections are washed, fixed in 10% formalin, embedded and cut

WO 94/06459

PCT/US93/08937

-35longitudinally to expose the morphology ox. the colon crypt. Tissue sections are coated with liquid emulsion and autoradiographed.

The proliferation index of cells in various stages of crypt development is determined by microscopic counting of exposed silver grains. Routinely, cells with &gt;4 grains score as positive. . Similarly cells labelled with anti-BrdU antibodies are counted as a percentage of the total number of cells’. In normal tissue, only the lower third of the crypt (containing the stem cell population) are labeled. Adjacent tissue serves as an internal control. The appearance of differentiation markers on colon crypt cells are monitored using available monoclonal antibodies to cytokeritins and colon specific antigen (fetal). Alternatively, cellularity of the villus and crypt may be used a over a period of 4 days as a marker of cell proliferation.

To establish in vitro models for the chemoprotective effects of Transforming Growth Factor Beta, doses of cytotoxic drug in organ culture or in dispersed mixed cell culture required for toxicity are assessed. Organ cultures are prepared as described previously (Shamsuddin, 1990) Parallel cultures are incubated in a range of 5-FU concentrations and proliferation measured by [3H]-thymidine incorporation and sectional autoradiography. To establish dispersed, mixed cell colon cultures, biopsy material are cut to &gt;0.5 cm², washed in phosphate buffered saline (PBS), finely titrated, centrifuged, rinsed, washed (5X) and cultured in a mixture of Leibowitz's medium L15 and suspension modified MEM (SMEM) with a final Ca⁺² concentration of 0.5mM, 10% fetal calf serum, 100 units penicillin, 50gg/ml streptomycin, 25gg/ml gentamicin, 2 mM glutamine, 1 ng/ml epidermal growth factor (EGF), 20 gg/ml insulin, 10 gg/ml transferrin, 25 nM sodium

WO 94/06459

PCT/US93/08937 ·

-36selenite and grown on collagen (Type I) coated culture plates (Wong et al., 1975, Moyer et al.,1990). Colon cells grown on collagen coated coverslips in a range of 5-FU concentrations are incubated in [.3H]-thymidine (0.2gCi/ml) for thirty minutes, washed and chased overnight in fresh media. Cells are washed, fixed/stained, coverslips dipped in liquid photographic emulsion and autoradiographed. Proliferation is measured by counting exposed silver grains. Chemoprotection by Transforming Growth Factor Beta in vitro is measured by pre-addition of a range of Transforming Growth Factor Beta doses to colon organ or cell cultures followed by 5-FU at or near the toxic dose, and [3H]-thymidine labelling at intervals after exposure to cytotoxic drug as a measure of cellular recovery.

The effective dose for systemic administration of Transforming Growth Factor Beta for gastrointestinal protection may be determined by animal modelling in mice. Histomorphic analysis of representative cross sections of small intestine at defined distances between the pylorus and the caecum are performed. Serial hematoxylin-eosin stained transverse sections of duodenum, jejunum and ilium were measured for numbers of crypts and villi per transverse section. The number of epithelial cells per representative villus section and the number of cells per representative crypt section are constant in the steady state and the rate of renewal is determined by the number of mitotic figures per representative crypt section. Kinetic analysis is obtained by autoradiography following ³H-thymidine or BrdU labelling in vivo (25/xCi ³H-thymidine/mouse or 5 mg BrdU/mouse). The labelling index of the tissue section provides a standard measure of the proliferation status of that tissue and demonstrates that Transforming Growth Factor Beta reversibly inhibits proliferation of the gastrointestinal epithelia.

WO 94/06459

PCT/US93/08937

-37Also, the proliferation of the gastrointestinal epithelia is assessed by measurement of villus and crypt cellularity, and villus height. Specifically, the small intestine is removed and the lumen flushed with cold (4°C) HBSS, the tissue is then briefly blotted dry and weighed. After fixation in Carnoy's fluid for 3 h, three segments (approx, 10% each of the total length) are removed from the proximal (duodenum), middle (ileum) and distal (ileum) regions of the small intestine, embedded in wax and cut as TS sections (4 gm thickness). Sections are stained with Hematoxylin and Eosin or PAS/Alcian blue (Goblet cells and mucins). Measurements are made of wet tissue weights; TS diameter of the intestine (gm, 10 measurements per animal); relative crypt: villus ratio (total number of crypts and villi per circumference - 5 sections of gut per animal); crypt height • in gm (measured from the base to the top of the crypt only in centrally sectioned crypts where there is an open lumen from the base of the top of the crypt, 20 measurements per animal); villus height (measured in gm from the crypt: villus junction to the villus tip, 20 measurements per animal; and the crypt cellularity (the number of epithelial nuclei measured from the base to the top of the crypt, but only in centrally sectioned crypts, 20 measurements per animal). These data allow determination of the preferred dose, schedule and route of administration for Transforming Growth Factor Beta mediated cell cycle inhibition.

In a typical experiment with Balb/c mice, all of the animals received two intravenous injections of 5-fluorouracil (5-FU) at 150 mg/kg body weight separated by seven days. Control mice were sacrificed one day after receiving the second intravenous 5-FU injection. TGF-B3 treated animals received four intraperitoneal (IP) injections of TGF-B3 (2gg per injection) around the time of the first 5-FU injection, five (IP) injections of TGF-B3 (2gg per injection) around the

WO 94/06459

PCT/US93/08937 ·

-38time of the second 5-FU injection and sacrificed one day following the second 5-FU injection. The first IP administrations of TGF-B3 were at 24 hours and 3 hours preceding, at the time of, and 3 hours following the first 5 intravenous 5-FU injection. The second administrations of

TGF-B3 were at 29 hours, 24 hours, and 3 hours preceding, at the time of, and 3 hours following the second 5-FU injection. Mice were sacrificed by cervical dislocation and the small intestine removed and immediately placed in a 10 phosphate buffered solution of 10% formalin (PBSF) at room temperature. The lumen was flushed with PBSF and the small intestine cut into three equal, segments as above. From each animal, five samples of 3 mm in length were cut from the length of each segment, transferred and stored in 5 ml of 15 PBSF. The samples were embedded in wax and transverse sections 5 μη in thickness were cut. The cut sections were transferred to slides and stained and measured as above.

As presented in Table 7, the length of the villus and crypt 20 from the animal receiving TGF-B3 was significantly longer than was noted in the animal receiving only the two injections of 5-FU. The crypt cellularity in the TGF-B3 treated animal was also greater than the animal that did not receive TGF-B3. Measurements were made from each section of 25 the number of cells in the crypt, the number of cells in the villus, the extent of apoptosis, villus height (in gm from trie crypt:villus junction) to trie villus tip.

Similar experiments are performed with TGF-B1, TGF-B2, TGF30 B4, and TGF-B5.

WO 94/06459

PCT/US93/08937

-392ΔΒ1Ε.7

<td></td><td> section</td><td> ViU.Us.Ljim)</td><td> Crvot(urn)</td><td> Crypt</td>

<td> Control</td><td> Duodenum</td><td> 15</td><td> 5</td><td> Cellularitv ND</td>

<td></td><td> Jejunum</td><td> 25</td><td> 5</td><td> 7</td>

<td></td><td> Ileum</td><td> 20</td><td> 5</td><td> 12</td>

<td> With TGF-B3</td><td> Duodenum</td><td> 55</td><td> 5</td><td> 20</td>

<td></td><td> Jejunum</td><td> 35</td><td> 10</td><td> 15</td>

<td></td><td> Ileum</td><td> 30</td><td> 10</td><td> 20</td>

ND=Not Determined

In another experiment this time with BDF-1 mice, all of the mice received two intravenous injections of 5-fluorouracil (5-FU) at 150 mg/kg body weight separated by four days. Control animals were sacrificed one, two, four and seven days following the second intravenous 5-FU injection. TGF-B3 treated animals received three intraperitoneal (IP) injections of TGF-B3 (5/ig per injection) around the time of the second 5-FU injection and sacrificed one, two, four and seven days after receiving the second 5-FU injection. The IP administrations of 5 ^g TGF-B3 were at 6 hours and 3 hours preceding and at the time of the second intravenous 5-FU injection. Mice were sacrificed by cervical dislocation and the small intestine removed and immediately treated as above.

As shown in Figure 10, the sections taken from the BDF-1 animal sacrificed four days following the second intravenous 5-FU injection and receiving Transforming Growth Factor Beta showed significantly greater number of villi and overall cellularity when compared with sections taken from an animal receiving only the two injections of 5-FU. Measurements were made from each section of the number of cells in the

WO 94/06459 PCT/US93/08937 ’

-40crypt, the number of cells in the villus, the extent of apoptcsis, villus height (in pm from the crypt:villus junction) to the villus tip.

In a third experiment, Balb/c mice received two intravenous injections of 5-fluorouracil (5-FU) at 150 mg/kg body weight separated by seven days.. Control animals were sacrificed one, two, four and seven days following the second intravenous 5-FU injection. Transforming Growth Factor Beta treated animals received 3 intraperitoneal (IP) injections of TGF-B3 in a similar manner to the BDF mice above. The mice were sacrificed and the intestine prepared as above.

As shown in Figure 11, the sections taken from the Balb/c animal sacrificed four days fallowing the second intravenous

5-FU injection and receiving TGF-B3 shows significantly greater number of villi and overall cellularity when compared with sections taken from an animal receiving the two injections of 5-FU without TGF-B3. Measurements were made from each section of the number of cells in the crypt, the number of cells in the villus, the extent of apoptosis, villus height (in Mm from the crypt:villus junction) to the villus tip. Thus, Transforming Growth Factor Beta increases crypt cellularity ar.J, villus length and prevents chemo-

5 induced apoptos i s,

Similar experiments are performed using TGF-Bi, TGF-B2, TGFB4, and TGF-B5 for in vivo growth inhibition of gastrointestir?;! epithelia.

Example. 7B: Use of Transforming Growth Factor Bata in

Protection of the Crypts of the Gl Tract from Irradiation

After irradiation of the G.I. tract some undamaged crypt35 base cells may be capable of regenerating crypts via clonal ‘ WO 94/06459

PCT/US93/08937

-41growth. Destruction of all clonogenic cells in a crypt results in loss )f that crypt within about 48 hrs. If sufficient crypts are lost the integrity of the mucosal liningis threatened, particularly in the region of the 5 villus. Breakdown in this integrity results in a loss of barrier function and the symptoms of the GI syndrome'ensue, which can result in death within the period 3-7 days postirradiation. The dose that causes death in 50% of the mice within 7 days for conventionally housed BDF1 mice, the 10 LD50/7, is about 11.1 Gy.

As described below, administration of Transforming growth factor B protects against crypt loss from irradiation and . thus ameliorates the severity of the GI syndrome which 15 ensues.

TGF-B3 was shewn to protect intestinal crypt cells from irradiation induced. damage. Mice were injected (ip) with TGF-B3 (2.5 gg) at -24, -8, -4 and 0 hours (Group B), or -4, 20 -3, -2, -1 and 0 hours (Group D) prior to irradiation at

1200 rads at time 0. Control groups (A and C) received saline injections in place of TGF-B3. Administration of TGF-B3 by both schedules enhanced survival of crypt cells in the gut.

.

WO 94/06459

PCT/US93/08937 *

<td colspan="4"> TGF-B3 CRYPT SURVIVAL (CLONOGENIC OR STEM CELL SURVIVAL)</td>

<td colspan="2"></td><td colspan="2"> Surviving crypts/ circumference at 3 days</td>

<td> Group A</td><td> -24H, -8, -4, 0 Saline injections</td><td> 37.3</td><td> Control</td>

<td> Group B</td><td> -24H, -8, -4, 0 TGF-B3 (2.5 gg) 1200 rads</td><td> 48.8</td><td> 31% Protection</td>

<td> Group C</td><td> -4H, -3, -2, -1, 0 Saline injections</td><td> 38.2</td><td> Control</td>

<td> Group D</td><td> -4H, -3, -2, -1, 0 TGF-B3 (2.5 gg) 1200 rads .</td><td> 50.9</td><td> 33% Protection</td>

Similar experiments are performed using TGF-Bl, TGF-B2, TGFB4, and TGF-B5 for in vivo growth inhibition of gastrointestinal epithelia.

Example 8? A Method for Prevention tat Effluvium or Alopecia Associated with Chemotherapy.

A common side effect of cytotoxic chemotherapy in the treatment of cancer is a thinning or complete loss of hair (effluvium or alopecia respectively). This side effect of cancer therapy is most common in patients receiving doxorubicin, but may also be observed in patients receiving methotrexate, cyclophosphamide, 5-fluorouracil, vincristine and other chemotherapeutic agents. This toxicity to the growing hair follicle is dose dependent. Approximately 8595% of hair follicles are actively growing, resulting in

WO 94/06459

PCT/US93/08937

-43scalp hair growth of -0.3 mm/day. Loss of hair during chemotherapy is psychologically distressful and has provoked some patients to discontinue therapy. The ability of Transforming Growth Factor Beta to inhibit cell cycling of many epithelial cell types is useful in reversibly slowing or blocking cycling of the hair follicle, thereby preventing chemotherapy associated hair loss. .

Several animals models have been developed for alopecia, in species where hair growth is asynchronous, as in humans. Species employed include the guinea pig and the Angora rabbit (Powis and Kooistra, 1987). The rabbit model is preferred. Rabbits are given 2 mg/kg doxorubicin, intravenously, twice a week for three weeks. Weekly measurements are made of: i) hair length; ii) shaved hair weight (obtained by shaving a 4 x 6 cm flank patch) ; iii) groomed hair weight (obtained by brushing) ; and iv) body weight. Transforming Growth Factor Beta is administered at a concentration of 1 Mg/kg to 250 Mg/kg, intraperitoneally, twice daily for up to three days prior to chemotherapy and may continue for up to 2 days after chemotherapy. Alternatively, Transforming Growth Factor Beta may be administered topically to the flank of the rabbit in a concentration range of 10 ng/cm² to 1000 pg/cxa²., twice daily for up to three days prior to chemotherapy and may continue for up to 2 days after chemotherapy. Protection by Transforming Growth Factor Beta Of chemotherapy associated hair loss is shown by increased hair length, increased shaved hair weight and increased groomed hair weight.

Based on animal modelling data, patients receiving chemotherapy may be administered Transforming Growth Factor Beta either systemically by intravenous injection at a concentration of 1 Mg/kg to 250 Mg/kg, and twice daily for up to three days prior to chemotherapy and may continue for

WO 94/06459

PCT7US93/08937 ‘

-44up to 2 days after chemotherapy. Alternatively, Transforming Growth Factor Beta may. be administered topically to the scalp in a concentration range of 10 ng/cm2 to 1000 Mg/cm2, twice daily for up to three days prior to chemotherapy and may continue for up to 2 days after chemotherapy. Transforming Growth Factor Beta may also be used to increase new blood vessel formation in the scalp by a transdermal delivery system or by subderma1 application using suitable penetration enhancers or by subcutaneous injection or other appropriate routes which will be recognized by those of skill in the art.

Example 9; Transforming Growth Factor Beta in vivo: Chemoprotection of Bone Marrow Cells

The effects of TGF-β3 on the murine hematopoietic system have been investigated in vivo and in vitro. Chemoprotection of bone marrow stem cells in vitro may be shown by the following method. The hematopoietic stem cells of Balb/C mice may be induced to cycle by treatment with 5-FU (150 mg/kg) . After four days the bone marrow is removed and treated with TGF-B3 and a chemotherapeutic agent including but not limited to doxorubicin, Ar-a-C, 5-FU, daunorubicin, methotrexate, vinblastine, vincristine and cyclophosphamide or with chemotherapy agent alone. Chemoprotection by TGF-B3 is measured by'culturing the so treated marrow cells in the presence of early acting cytokines (for example kit ligand, IL-1, IL-3) or conditioned medias which support the proliferation of primitive, pluripotent stem cells (for example Mo cell conditioned media). Protected cells will proliferate to form large colonies characteristic of self-renewing pluripotent stem cells. The measure of protection is the increased number of high proliferative potential (HPP) colonies in the TGF-B3 treated versus the untreated marrow. Low proliferative potential (LPP)

WO 94/06459

PCT/US93/08937

-45colonies are smaller more committed hematopoeitic cells. An example of data obtained with in vitro chemoprotection is below:

In Vitro TGF-B3 Chemoprotection of 5FU Enriched Mouse Marrow

<td></td><td> HPP/LPP</td><td> Percent of Control</td>

<td> Control</td><td> 687/9983</td><td> 100%/100%</td>

<td> TGF-B3 only</td><td> 81/694</td><td> 11.8%/6.9%</td>

<td> Ara C only</td><td> 1/7</td><td> 0.14%/0.07%</td>

<td> TGF-B3 and Ara C</td><td> 26/158</td><td> 3.8%/1.6%</td>

Chemoprotection by TGF-B3 in vivo is shown by the following method. Balb/C mice are exposed to a cytotoxic agent (typically 5-FU or ara-C) to induce the pluripotent HPP stem cells into cell cycle. After a suitable time period (usually 4 days) a second dose of cytotoxic agent is administered to the animal. Prior to the second dose of cytotoxic agent, TGF-B3 is administered to the animal for a suitable time (generally 2 gg, twice daily for one to three days). A suitable time after the second dose of cytotoxic agent is administered (generally 2 to 4 days) , stem cell number and proliferative status is measured. In Transforming Growth Factor Beta treated mice, increased numbers of stem cells are present following chemotherapy .(for example 5-FU or ara-C) compared with untreated mice. Similarly, in vivo chemoprotection of marrow stem cells may be observed by the following protocol. TGF-B3 (2 gg, intraperitoneal) is administered to Balb/C mice 4 hours prior to 5-FU (IV, 150 mg/kg) and TGF-B3 treatment continued (2 gg, intraperitoneal) at the time of injection, 4 hours post 5-FU and 20 hours post 5-FU. Marrow is isolated, cultured and stem cell number quantitated. Treatment with Transforming Growth Factor Beta results in a four fold greater number of surviving stem cells with which to reconstitute the blood system. Transplantation of lethally irradiated Balb/C mice

WO 94/06459

PCT/US93/08937 ·

I

-46(850 rads) with Transforming Growth Factor Beta protected stem cells can either fully reconstitute or lead to significantly increased life spans of transplanted mice, depending on the protocol employed. In an additional example, six Balb/c mice received two intravenous injections of 5-fluorouracil (5-FU) separated by four days. The first injection of 5-FU was at 150 mg/kg and the second injection of 5-FU was at 130 mg/kg. Three control mice were sacrificed one day after receiving the second intravenous 5FU injection. Three TGF-B3 treated animals received three intraperitoneal (IP) injections of TGF-B3 (2gg per injection) . The three IP administrations of 2 μ-g TGF-B3 each were at 6 hours and 3 hours preceding and at the time of the second intravenous 5-FU injection. All mice were sacrificed by cervical dislocation, femoral bone marrow removed and colony forming cells assayed using a doublelayer agarose system. Sixty-millimeter petri dishes containing a 2-ml underlayer consisting of culture media, cytokines and 0.5% agarose were overlayed with 1 ml of BM cells suspended in culture media containing 0.36% agarose. The final concentration of cytokines in the double-layer were 10% WEHI-3-conditioned medium (Shieh, J.-H. et al., 1991) (as a source of murine IL-3), 100 U/ml IL-1B and 1000 U/ml mGM-CSF. The cell concentration in the 1 ml cell layer was 1 X 10⁵ cells. Double layer cultures were grown for 10 days at 37° C under low oxygen tension (5%0₂/5%C0₂/90%N₂). Each 60 mm plate was either fixed with 2 ml of 0.5% glutaraldehyde in PBS or stained with 2 ml of a solution of iodonitrotetrazolium (INT) (1 mg/ml) and scored for colony formation. HPP and LPP colony classifications were as previously described (Moore, et al., 1987).

WO 94/06459 PCT/US93/08937

-47- '

In Vivo Chemoprotection of Bone Marrow Cells by Transforming Growth Factor Beta

Average colonies per mouse

<td> 5</td><td></td><td> HPP</td><td> LPP</td><td> TOTAL</td>

<td></td><td> Control Unprotected</td><td> 542</td><td> 2396</td><td> 2938</td>

<td></td><td> TGF-B3 Chemoprotected</td><td> 949</td><td> 4129</td><td> 5079</td>

In a similar manner, experiments are performed with TGF-Bl, 10 TGF-B2, TGF-B4, and TGF-B5.

Treatment of patients with Transforming Growth Factor prior to multicycle cytotoxic chemotherapy, will protect against hematopoietic stem cell loss and thereby reduce the 15 incidence of febrile neutropenia, the number of patient hospital days, reduce the use of antibiotics, and allow successful completion of chemotherapy.

Example 10: Transforming Growth Factor Beta In Vivo:

Effects on Survival without Cytokines and Synergy with Cytokines ·

Several strategies can be employed for in vivo administration of TGF-B3 as a hematopoietic chemoprotectant. 25 TGF-B3 is used during or immediately prior to chemotherapy to protect the early hematopoietic stem cell population in multicycle or accelerated cycle chemotherapy. Hematopoietic growth factors can be successfully used in the recovery from myelosuppression. Prevention of long term stem cell 30 depletion, neutropenia and continued tolerance to further chemotherapy are clinical endpoints measured in trial.

(a) Protection from the effects of 5-FU in the absence of cytokines:

WO 94/06459

PCT/US93/08937 -48Two groups of Balb/c mice were treated with 150 mg/kg 5-FU administered i.v. on day zero followed by 130 mg/kg on day

4. The TGF-B3 treated group received five doses of 2 μg TGF-B3 each in PBS administered by injection i.p. The first dose was given on day 3 and the remaining four doses were given 3 hours prior to, coincident with, 3 hours after and one day after the second 5-FU treatment. Control animals received injections of PBS alone.

The survival results on day 9 were as follows:

Survival TGF-B3 treated group 40%

Control Group 0%

These results demonstrate that TGF-B3 is capable of enhancing survival of 5-FU treated mice in the absence of exogenously added cytokines. · (b) Protection from the effects of 5-FU in the presence of exogenously administered cytokines: ·

For example, Balb/C mice were treated with two cycles of 5FU (150 mg/kg) spaced seven days apart, which is normally a lethal course of chemotherapy in these animals. TGF-B3 administration significantly increased survival time (p=.0009). percent survival is shown for day 26 post 5-FU (Figure 12A).

WO 94/06459

PCI7US93/08937

<td> 2 X 5-FU [n=12]</td><td> 8% survival</td>

<td> 2 X 5-FU + cytokine rescue [n=10]</td><td> 0%</td>

<td> 2 X 5-FU + TGF-B3 (3X2 pg; 6hrs.)</td><td></td>

<td> + cytokine rescue [n~5]</td><td> 50%</td>

<td> 2 X 5-FU + TGF-B3 (4X2 pg; 24hrs.)</td><td></td>

<td> + cytokine rescue [n=10]</td><td> • 50%</td>

<td> 2 X 5-FU + TGF-B3 (6 X 2 pg; 30 hrs.)</td><td></td>

<td> + cytokine rescue [n=5]</td><td> 15%</td>

Figure Ϊ2Β shows weight retention data for the same experiment. Note that very high doses of TGF-B3 and 5-FU accelerates weight loss and mortality.

A similar survival study also showed increased survival and weight retention in TGF-B3 treated mice (day 17 post 5-FU; (Figure 12C):

X 5-FU 0% survival

X 5-FU + cytokine rescue (G-CSF + IL-1) 20%

X 5-FU + TGF-B3 (5 X 2 pg; 27hrs.) + cytokines 80%

Alternatively, hematopoietic growth factors (eg. kit ligand) may be administered before chemotherapy to increase total bone marrow cellularity and increase the number of hematopoietic cells surviving cytotoxic chemotherapy. Expanding the hematopoietic stem cell population using growth factors requires a mechanism to restore these cycling cells back to a non-dividing state prior to chemotherapy. TGF-B is unique in its ability to perform this function. TGF-B3 is used to protect this expanded population of hematopoietic stem cells prior to chemotherapy and the increased reserve of the more mature hematopoietic cells reduces the fall in white blood cell count associated with chemotherapy.

WO 94/06459

PCT/US93/08937 «

-50For example we have shown TGF-B3 can protect BDF1 mice from the lethal effects of stem cell factor (SCF) and chemotherapy (5-FU). In this model SCF is administered prior to 5-FU treatment to deplete the stem cell population.

<td colspan="5"> As shown below this SCF/5-FU combination is lethal and is</td>

<td> used here to measure</td><td colspan="2"> TGF-B3 mediated</td><td> stem</td><td> i cell</td>

<td colspan="2"> chemoprotection. Two cycles of 5-FU,</td><td> SCF</td><td> and</td><td> TGF-B3</td>

<td> treatment were used, spaced</td><td> 7 days apart.</td><td> Day</td><td colspan="2"> 17 survival</td>

<td> data and day 10 blood cell</td><td> counts (WBC)</td><td colspan="2"> are shown</td><td> (Table</td>

<td> 10) .</td><td></td><td></td><td></td><td></td>

<td> TABLE' 10</td><td></td><td></td><td></td><td></td>

<td> Treatment</td><td> Survival</td><td></td><td> WBC</td><td> (io⁶)</td>

<td> 5-FU (n=10) ·</td><td> 100%</td><td> -</td><td> 12.6</td><td></td>

<td> 5-FU + TGF-B3 (n=10)</td><td> 100%</td><td></td><td> 15.9</td><td></td>

<td> 5-FU + SCF (n=10)</td><td> 0%</td><td></td><td> 6.0</td><td></td>

<td> 5-FU + TGF-B3 + SCF (n=10)</td><td> 40%</td><td></td><td> 6.1</td><td></td>

Similar in vivo experiments are performed using TGF-fil, TGFB2, TGF-B4, and TGF-B5.

Example 11; Acute Toxicology and Chemoprotection of Normal Mice

Acute administration of escalating doses of Transforming Growth Factor Beta can be investigated for toxicity (survival, weight loss) in normal Balb/c or BDF1 mice. Various hematologic parameters, progenitor cell assays and immune function assays are undertaken. The status of the hematopoietic stem cell compartment (total numbers in marrow, spleen and circulation, and cell cycle status) is determined using the in vivo CFU-S assay, and the in vitro CFU-GEMM and high proliferative potential (HPP) CFU assays. Progenitor cells for the erythroid (BFU-E, CFU-E), myeloid (CFU-GM, CFU-M, CFU-G) and megakaryocyte/platelet series (CFU-MK) are assayed. Lymphoid function is measured by BWO 94/06459

PCT/US93/08937

-51and pre B-lymphocyte colony assays, etc. Subsets of myeloid and lymphoid cells in tissues is determined by FACS analysis using lineage-specific monoclonal antibodies. White blood cells (WBC), platelets, red blood cells (RBC) and hematocrit are measured in repeated tail vein blood samples. Serum samples are obtained and are assayed for TGF-β, G-CSF, GM-CSF, M-CSF, IL-1,3,4,5 and TNF by bioassay and radioimmunoassay. Neutrophil function assays include in' vivo and in vitro chemotaxis, bactericidal capacity and receptor expression for multiple cytokines.

The ability of Transforming Growth Factor Beta to protect hematopoietic stem cells from the cytotoxic effects of chemotherapeutic agents is assessed in animal models. Untreated BAiB/c mice, or groups pre-treated with Transforming Growth Factor Beta,, receive a single i.v. dose of 5-FU (150 mg/kg). In subsequent experiments cyclophosphamide (200 mg/kg), vinblastine (2.5 mg/kg), adriamycin (2.5 mg/kg) or methotrexate (150 mg/kg) is employed. The number of peripheral circulating blood cells and the various hematopoietic progenitor cells are determined, including CFU-S, CFU-GM, HPP-CFU-C, BFU-E, CFU-Mk and CFU-GEMM.

It is preferable, although not necessary, to identify the optimal parameters -of dosage and scheduling relationships by the experiments hereinabove, i.e. Transforming Growth Factor Beta is tested in this model. Most of the planned studies are performed using first generation syngeneic transplants, of spontaneous breast tumors. In addition to determination of life span, the effect of therapy is determined on tumor growth rate, on partial and complete regression, and on spontaneous metastasis (determined histologically or by tissue retransplantation). Transforming Growth Factor Beta is tested for its ability to mediate a reversible cytostatic

WO 94/06459

PCT/US93/08937

-52block on hematopoietic progenitor and stem cell proliferation, conferring resistance to toxicity of chemotherapy. ·

Similar in vivo experiments are performed using TGF-Bl, TGF52, TGF-54, and TGF-55.

Example 12: Chamoprotection of Primary Human Bone Marrow and Peripheral Blood Samples

Primary human bone marrow and peripheral blood samples were evaluated for the growth of the different hematopoietic ’ lineages in the presence and absence of exogenous

Transforming Growth Factor Beta. Stem cell cultures were purged of mature, well differentiated cells. Specifically, buffy coat cells were collected, by centrifugation (800 rpm, 10', 5°C), suspended in McCoy's medium including 10% heat inactivated FCS ('complete medium'). Platelets were removed by Percoll gradient centrifugation (1.050 g/ml; Pharmacia) and a low density, small primitive cell population obtained after re-centrifugation on a Percoll gradient (1.075 g/ml). Individual populations of B- and T-lymphocytes, granulocytes, monocytes and more differentiated erythroid populations can be immunodepleted (Centrella et al., 1986) by panning 2xl0⁸ light-density cells with monoclonal antibodies (anti-Bl, anti-B4, anti-LyT3, anti-My4, anti-My8, anti-903, anti-N901, anti-Leul and anti-glycophorin A (R10) and WEM-Gll) directed against mature cell surface epitopes for 30 minutes on ice. Cells were washed twice in cold complete medium and assayed for progenitor cells (Strife et al., 1988, Strife, 1987). The primitive stem cell population were grown in methylcellulose support (Iscove's modified Dulbecco's medium (IMDM; Gibco), 24% FCS, 0.8% dialyzed bovine serum albumin, ΙΟΟμΜ β-mercaptoethanol, and 1.3% methylcellulose in 35mm Lux culture plates) in the

WO 94/06459

PCiT/US93/08937

-53presence of 10 pM, 100 pM and 1000 pM TGF-B3 with and without Mo T-lymphocyte conditioned media, a source of colony stimulating factors (Golde, D.W. et al., 1980) in quadruplicate cultures maintained in 5% C0₂.

Colony forming units of the various hematopoietic lineages were counted microscopically at 3, 7, 14 and 21 days. An example of this experiment is shown in Figure 13. Individual colonies were aspirated onto glass slides, 10 selectively stained with May-Grunewald-Giemsa and the presence of neutrophils, monocytes or eosinophils observed. These experiments allow one skilled in the art to determine if the growth inhibitory action of Transforming Growth Factor Beta is lineage-specific at given doses of 15 Transforming Growth Factor Beta, determine the time course of inhibition and determine the dose of Transforming Growth Factor Beta required for inhibition of a given cell type.

Figure 14 shows a theoretical plot of TGF-B3 chemoprotection 20 to reduce chemotherapy-induced reduction of normal blood count. As a general rule, the smaller hematopoietic precursors represent more primitive progenitor stem cells while the larger cells are usually more mature, as analyzed by the appearance of maturation-specific cell surface 25 epitopes. Enriched progenitor populations obtained by immuno-depletion as described may be size selected by Percoll gradient centrifugation and different size cell populations evaluated for specific lineages in combinations of Mo conditioned media and TGF-B3 at 3, 7, 14 and 21 days.

Assays for early stem, .cell populations-(HPP-CFU of CFU-A), progenitor cells (CFU-E, CFU-GEMM, BFU-E, CFU-MK), pre-B colony, B colony, T colony, cytolytic T cell, and antigen stimulation assays are currently well developed.

WO 94/06459

VCT/JS93/08937 &lt;

*

-54Example 13 ϊ Use of Transforming Growth Factor Beta in Autologous Bone Marrow Transplantation and Syngeneic Mouse Model

Autologous bone marrow transplantation is a method in which the bone marrow of a patient is removed prior to chemotherapy to reduce hematopoietic stem cell toxicity. Autologous bone marrow transplantation has also been performed in patients with acute nonlymphocytic leukemia using 4-hydroperoxycyclophosphamide to treat the marrow ex vivo (Cashman, J.D. et al., 1990).

In one instance, Transforming Growth Factor Beta is used to inhibit the proliferation of the bone marrow stem cell population prior to chemotherapy of bone marrow in vivo in patients with blood-bone tumor cells. Briefly, Transforming Growth Factor Bota is contacted with the patient bone marrow at a sufficient concentration to inhibit the normal hematopoietic cells (for example, 1-1000 pM) as determined by one skilled in the art. At a given time after Transforming Growth Factor Beta treatment of bone marrow ex vivo (typically but not limited to 6-24 hours or as determined by the physician) and the tumor cell population refractory to the effects of Transforming Growth Factor Beta, th® cells are treated with a cytotoxic chemotherapeutic agent (e.g. adriamycin, 5-fluorouracil, and vinblastine). The treated marrow is returned to the patient at a time determined by the physician and the normal hematopoietic cells allowed to recover from the growth inhibiting effects of .Transforming Growth Factor Beta and proliferate, thus reconstituting that component of the patients hematopoietic system.

Alternatively, the bone marrow cells are treated with Transforming Growth Factor Beta as described and the bone

WO 94/06459

PCT/US93/08937

-55marrow cells cultured ex vivo as described such that the leukemic cell population continues to proliferate and terminally differentiate while the normal population is growth arrested. Continued culture in this way results in the terminal growth arrest of the leukemic population and enrichment of the normal cell population. Further, enrichment of the normal cell population may .be accelerated by contacting the normal cell population with hematopoietic growth factors, e.g. kit ligand, G-CSF, GM-CSF, M-CSF, IL-1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11. Bone marrow thus treated is returned to the patient, substantially free of leukemic cells.

A bone marrow transplantation experiment in a syngeneic Balb/C mouse model was performed as follows: TGF-B3 (10 ' ng/'ml) treated bone marrow cells taken from donor mice were protected from the cytotoxic effects of Ara-C (100 gg/ml) and this bone marrow population was then- transplanted into lethally irradiated mice, enhancing their long term (greater than 25 day) survival (20% survivors). Control animals transplanted with bone marrow cells which had not been protected by TGF-B3 treatment before Ara-C exposure resulted in no survivors (0% survival).

Similar ex vivo hematopoietic chemoprotection and in vivo transplantation experiments are performed using TGF-Bl, TGFB2, TGF-B4, and TGF-B5.

Example 14? Demonstration of Time and Concentration Dependent Chemoprotection by Transforming Growth Factor Beta of Early Hematopoietic Stem cells from Ara-C In Vitro.

In this experiment, femoral bone marrow enriched for early hematopoietic progenitors were taken from Balb/c mice 4 days after receiving a single intravenous injection of

WO 94/06459

PCT/US93/08937 ·

-565—fluorouracil (5-FU) at 150 mg/kg body weight. Harvested bone marrow cells (BM) was diluted to 7.5 xlO⁵ cells/ml

I with culture media consisting of Iscove s modified

Dulbecco* s medium with 20% fetal bovine serum (FBS) supplemented with 100 U/ml purified recombinant human IL-1B and 20 ng/ml purified recombinant murine kit ligand (mKL) and 1 ml was added to each well of a 24-well cluster plate. Starting with the beginning of the third day, TGF-B3 was added at 10 ng/ml final concentration at twenty-four, twelve, and zero hours prior to the addition of Ara-c and incubated at 37°C in a fully humidified 5% CO₂ atmosphere.

Ara-C at a final concentration of 100 Mg/ml was added at the beginning of the fourth day and incubated for an additional 24 hours at 37°C in a fully humidified 5% CO₂ atmosphere.

At the end of the incubation, the various treated BM was diluted once (1:10) and washed three times by centrifugation

I with chilled Dulbecco s phosphate buffered saline containing

10% FBS at ;°C. Colony forming cells were assayed using a double-layer agarose system. Sixty-millimeter petri dishes 20 containing a 2-ml underlayer consisting of culture media, cytokines and 0.5% agarose were overlayed with 1 ml of treated and washed BM cells suspended in culture media containing 0.36% agarose. The final concentration of cytokines in the double-layer was 10% WEHI-3-conditioned 25 medium, as a source of murine IL-3, (Sheih et al., 1991),

100 U/ml IL-ΐβ and 1000 U/ml rmGM-CSF. The cell concentration in the 1 ml of cell layer was 1 x 10⁴ cells of control cells or 1 x 10⁵ cells treated with Ara-C, with or without TGF-B3. Double layer cultures were incubated for 9 30 days at 37°C under low oxygen tension (5%0₂/5%C0₂/90%N₂).

Each 60 mm plate was fixed with 2 ml of 0.5% glutaraldehyde in PBS and scored for colony formation.

Figure 15 demonstrates Transforming Growth Factor Beta 35 time-dependent chemoprotection of the early hematopoietic ’ WO 94/06459

PCT/US93/08937

-57stem cells from the cytotoxic effects of chemotherapeutic agents (i.e. Ara-C) . Cells treated simultaneously with TGF-B3 and Ara-C shows significant chemoprotection relative to cells treated with Ara-C only. Longer pretreatment of cells with TGF-B3 (i.e. 12 hrs), prior to exposure with Ara-C, demonstrated a greater level of protection than when cells were treated with TGF-B3 sx_ iltaneously with Ara-C. Pretreatment of cells with TGF-B3 24 hrs prior to exposure to Ara-C continues to yield significant chemoprotection. Since TGF-B3 is growth inhibitory, fewer colonies are observed by the group pretreated for 24 hrs with TGF-B3 when compared to the group pretreated for 12 hrs due to the presence of fewer cells at .the time of Ara-C exposure and subsequent processing.

In another experiment, femoral bone marrow enriched for early hematopoietic progenitors were taken from Balb/c mice 4 days after receiving a single intravenous injection of 5-fluorouracil (5-FU) at 150 mg/kg body weight. Harvested bone marrow cells (BM) at 3.75 xio⁵ cells in 1 ml were

I incubated in culture media consisting of Iscove s modified Dulbecco’s medium with 20% fetal bovine serum (FBS) in 24-well cluster plates supplemented with 100 U/ml purified recombinant human IL-1B and 20 ng/ml purified recombinant murine kit ligand (mKL, Muench et al., 1992) . At the end of the fourth day in liquid culture, TGF-B3 was added at the indicated concentration (0.001 ng/ml to 100 rig/ml final concentration) and further incubated for an additional fourteen hours at 37°C in a fully humidified 5% C0₂atmosphere. Following the fourteen hour incubation with TGF-B3, Ara-C at a final concentration of 100 gg/ml was added and incubated for an additional 28 hours at 37°C in a fully humidified 5% C0₂ atmosphere. At the end of the incubation, the various treated BM was diluted once (1:10) and washed three times by centrifugation with chilled

WO 94/06459

PCT/US93/08937 · t

-58*

Dulbecco s phosphate buffered saline containing 10% FBS at 4°C. Colony forming cells were assayed using a double-layer agarose system. Sixty-millimeter petri dishes containing a

2-ml underlayer consisting of culture media, cytokines and 0.5% agarose were overlayed with 1 ml of treated and washed BM cells suspended in culture media containing 0.36% agarose. The final concentration of cytokines in the double-layer was 10% WEHI-3-conditioned medium 1 (as a source of murine IL-3), 100 U/ml IL-IB and 1000 U/ml rmGM-CSF. The cell concentration in the 1 ml of cell layer was 4 x 10⁴ cells treated. Double layer cultures were incubated for 10 days at 37°C under low oxygen tension (5%0₂/5%C0₂/90%N₂). Each 60 mm plate was either fixed with 2 ml of 0.5% glutaraldehyde in PBS or stained with 2 ml of a solution of iodonitrotetrazolium (INT) (1 mg/ml) and scored for colony formation. Figure 16 demonstrates Transforming Growth Factor Beta concentration dependent chemoprotection of the early hematopoietic stem cells from the cytotoxic effects of chemotherapeutic agents (i.e. Ara-C). There is a significantly greater number of HPP cells and total colony forming cells when pretreated with &gt;0.01 ng/ml TGF-B3 relative to the control cells exposed to Ara-C without TGF-B3. At and above 1 ng/ml TGF-B3 shows similar levels of chemoprotection by TGF-B3 . There does not appear to be less chemoprotection when higher concentrations of TGF-B3 are used.

Similar ex vivo hematopoietic chemoprotection experiments are performed using TGF-Bl, TGF-B2, TGF-B4, and TGF-B5.

Example 15: Transforming Growth Factor Beta for Protection of Lung Tissue

TGF-B3 is used as a radioprotectant and chemoprotectant of normal lung tissue. Radiation induced and chemotherapy

WO 94/06459

PCT/US93/08937

-59induced pulmonary cytotoxicity is reduced by administration of any one of the following TGF-Bs: TGF-Bl, TGF-B2, TGF-B3, TGF-B4 or TGF-B5. The TGF-B is preferably administered by inhalation (e.g. of a vapor, mist or aerosol), but systemic 5 delivery is also effective. TGF-B is preferably administered prior to chemotherapy or radiotherapy.' TGF-B prevents cytotoxic poisoning of normal lung epithelial cells and normal lung endothelial cells.

WO 94/06459

PCT/US93/08937 ·

-60TABLE A (Additional chemotherapeutic drugs)

Gilman et al., 1990

2-Ethylhydrazide

2,2',2-Trichlorotriethylamine

6-Diazo-5-oxo-L-norleucine Aceglatone

Aclacinomycins Actinomycin F_xAminoglutethimide Amsacrine

Ancitabine

Anthramycin

Azacitidine

Azaserine

Benzodepa Bestrabucil

Bisantrene

Bleomycin

Buserelin

Busulfan Cactinomycin

Calusterone

Cai'boplatin Carboquone Carmofur

Carmustine (BCNU)

Carubicin

Chlorambucil Chlormadinone Acetate Chlornaphazine Chlorozotocin

Chromomycins

Cisplatin (cis-DDP) Cyclophosphamide Cytarabine (cytosine arabinoside; Ara-C)

Dacarbazine (DTIC; dimethyltriazenoimidazolecarboxamide)

Dactinomycin (actinomycin D)

Daunorubicin (daunomycin; rubidomycin)

Defosfamide

Demecolcine

Denopterin

Diaziquone Diethylstilbestrol Doxifluridine

Doxorubicin

Dromostanolone Propionate Eflornithine

Elliptinium Acetate

Enocitabine

Epirubicin

Epitiostanol

Estramustine

Ethinyl estradiol

Etoglucid

Etoposide Floxuridine (fluorodeoxyur idinfi; Fudr)

WO 94/06459

PCT/US93/08937

<td colspan="3"> -61-</td>

<td></td><td> Fludarabine</td><td> mercaptopurine; 6-MP)</td>

<td></td><td> Fluorouracil (5-</td><td> Methotrexate (amethopterin)</td>

<td></td><td> fluororuracil; 5-FU)</td><td> Meturedepa ·</td>

<td></td><td> Fluoxymesterone</td><td> Mitguazone</td>

<td> 5</td><td> Flutamide</td><td> Mitobronitol</td>

<td></td><td> Fosfestrol</td><td> Mitolactol</td>

<td></td><td> Fotemusine</td><td> Mitomycin (mitomycin C)</td>

<td></td><td> Gallium Nitrate</td><td> Mitotane (ο,ρ'-DDD) ’</td>

<td></td><td> Goserelin</td><td> Mitoxantrone</td>

<td> 10</td><td> Hexamethylmelamine</td><td> Mopidamol</td>

<td></td><td> Hexestrol</td><td> Mycophenolic Acid</td>

<td></td><td> Hydroxyprogesterone caproate</td><td> Nilutamide</td>

<td></td><td> Hydroxyurea</td><td> Nimustine</td>

<td></td><td> Ifosfamide</td><td> Nitracrine</td>

<td> 15</td><td> Improsulfan ’</td><td> Nogalamycin</td>

<td></td><td> Interferon alfa ·</td><td> Novembichin</td>

<td></td><td> Interferon-gamma</td><td> Olivomycins</td>

<td></td><td> Interferon-B</td><td> Pentostatin (2'-</td>

<td></td><td> Interleukin-2</td><td> deoxycoformycin)</td>

<td> 20</td><td> L-Asparaginase</td><td> Peplomycin</td>

<td></td><td> Lentinan</td><td> Phenamet</td>

<td></td><td> Leuprolide</td><td> Phenesterine</td>

<td></td><td> Leuprolide</td><td> Pipobroman</td>

<td></td><td> Lomustine (CCNU)</td><td> Piposulfan</td>

<td> 25</td><td> Lonidamine</td><td> Pirarubicin</td>

<td></td><td> Mannomustine</td><td> Plicamy’cin (mithramycin)</td>

<td></td><td> Mechlorethamine (HN₂)</td><td> Podophyllinic Acid</td>

<td></td><td> Meehlorethamine Oxide</td><td> Polyestradiol Phosphate</td>

<td></td><td> Hydrochloride ’</td><td> Porfiromycin ·</td>

<td> 30</td><td> Medroxyprogesterone acetate</td><td> Prednimustine</td>

<td></td><td> Megestrol acetate</td><td> Prednisone</td>

<td></td><td> Melengestrol</td><td> Procarbazine (N-</td>

<td></td><td> Melphalan (L-sarcolysin)</td><td> methylhydraz ine, MIH)</td>

<td></td><td> Mepitiostane</td><td> PSK® (KRESTIN)</td>

<td> 35</td><td> Mercaptopurine (6-</td><td> Pteropterin</td>

WO 94/06459

PCT/US93/08937 -62Puromycin

Ranimustine

Razoxane

Semustine (methyl-CCNU)

Sizofiran

Spirogermanium

Streptonigrin

Streptozocin (streptozotocin)

Tamoxifen

Taxol

Teniposide

Tenuazonic Acid

Testolactone

Testosterone propionate

Thiamiprine

Thioguanine (6-thioguanine;

TG)

Thiotepa

Toremifene

Traziquone

Tr i ethylenemelamine

Triethylenephosphoramide Trilostane

Trimethylolmelamine

Trimetrexate

Triptorelin

Trofosfamide

Tubercidin

Ubenimex

Uracil Mustard

Uredepa

Urethan

Vinblastine (VLB)

Vincristine Vindesine Zinostatin Zorubicin

WO 94/06459

PCT/US93/08937

-63 —

REFERENCES

1.

2.

3.

4.

5.

6.

7.

8.

9.

10.

11.

12.

13.

14.

15.

16.

17.

18.

19.

20.

Assoian et al. (1983) J. Biol. Chem, 258: 7155 «runner et al. (1988) Mol- Cell. Biol. 8(5):2229. Cashman, J.D. et al. (1990) Blood, 75:96.

Qentrella, et al. (1987) J. Biol. Chem, 262:2869.

Centrella, et al. (1986) Endocrinol. 119:2306. Cerletti, N. et al., European Patent .Publication No. 0 433,225 Al, published June 19, 1991.

Derynck. R. et al., U.S. Patent No. 4,886,747, issued December 12, 1990.

Eisen, D. et al. (1990) J. Amer. Acad, of

Dermatology 23:1259-1264.

Gesme (1992) Contemporary Oncology 2:35.

Gilman, A.G. et al., Eds. (1990) The

Pharmacological Basis of Therapeutics, 8th Ed.. pp. 1205-1207, Pergamon Press, New York.

Goey, H. et al.(1989) J. Immunol. 143:877-880.

Golde, D.W, et al. (1980) Proc. Natl, Acad, Sci. USA 77, 593-597.

Guggenheimer, J. et al. (1977) Oral Surg. Med. Oral Pathol♦, 44:58-63.

Hornung and Longo (1992) Blood 80: 77-83

Iwata, K.K. et al. (1985) Cancer Res. 45:2689-2694. Iwata, K. K., et al., (1992 A) International Publication No. WO 92/00318, published January 9, 1992.

Iwata, K. K., et al., (1992 B) International Publication No. WO 92/00330, published January 9, 1992.

Lockhart, O.B. and Sonis, s.T. (1981) J. Dermatol sure. 7:1019-25.

Madisen, L. et al. (1989) DNA 8:205-212.

Madisen et al. (1988) J. Cell Biochem. Suppl. pp 199.

WO 94/06459

PCT7US93/08937 ,

21. McMaster et al. EP 542 679 Novel Hybrid

Transforming Growth Factors

22. Moore, M.A.S. et al. (1987) PNAS 84:7134-7138.

23. Mossman, T. (1983) J. Immuno. Methods 65:55-65.

24. Moyer, M. et al. (1990) In Colon Cancer Cells. M.

Moyer and G. Poste, Eds. pp.85-136, Academic Press, New York. .

25. MUench, M.O. et al. (1992). Exp. Hematol, 20:339-349.

26. Nilsson, B. et al. (1988) In: Advances in Gene Technology; Protein Engineering and Production. K. Brew et al., Eds., IRL Press, Washington, D.C.:pp. 122-123.

27. Ogawa, Y. et al., (1992) J. Biol. Chem. 267:2325.

<td> 28.</td><td colspan="3"> Perlman and Halvorson (1983) J. Mol. 409.</td><td colspan="2"> Biol. 107:391-</td>

<td> 29.</td><td> Pincher</td><td> et al. (1985)</td><td> Biochem.</td><td> Biophvs.</td><td> Res,</td>

<td></td><td> Commun.</td><td> , 133(3):1026.</td><td></td><td></td><td></td>

<td> 30.</td><td> Posner,</td><td> M.R. (1990)</td><td> J. Natl.</td><td> Cancer</td><td> Inst.</td>

<td></td><td> 82(21):</td><td> 1710-1714.</td><td></td><td></td><td></td>

<td> 31.</td><td> Potten,</td><td> C.S. (1990) Int.</td><td> Radiat. Biol</td><td> .. 58:925-</td><td> 973</td>

<td> 32.</td><td> Potten,</td><td> C.S. (1991) in</td><td> Chemically</td><td> Induced</td><td> Cell</td>

Proliferation. Wiley-Liss, Inc., p.155-171

33. Purehio, A.F., et al., UK Patent Publication No. GB-

2210620A, published June 14, 1990

34. Shamsuddin, A.M. (1990) In': Colon Cancer Cells. M.

Moyer and G. Poste, Eds., pp 137-153, Academic Press, New York.

35. Shieh, J.-H., Peterson, R.H.F., and Moore, M.A.S. (1991). J. Immunol. 147:2984-2990.

36. Sonis, S. et al. (1978) JADA. 97:468-472.

37. Sonis, A. and Sonis, S. (1979) Journal of Pedodontics Winter:122-128.

38. Sonis, S. et al. (1988) Oral Sure. 65:19-22.

WO 94/06459

PCT/US93/08937

39. Sonis, S. et al. (1990) oral sura. Oral Med, Oral Pathol. 691437-443.

40. Sonis, S., (1990) In: gancer Principles and Practice or Oncology. Ed. deVita, V.; Hellman, S.; Rosenberg,

S.J. Volume 2, 3rd Edition. J.B. Lippincott

Company, Philadelphia.

41. Stover, D.E., (1990) In: Cancer Principles and Practice or Oncology. Ed. deVita, V.; Hellman, S.; Rosenberg, S.J. Volume 2, 3rd Edition. J.B.

Lippincott Company, Philadelphia.

42. Sonis, S., and Clark, J. (1991) Oncology. 5(12):11-

17.

43. Strife, A. et al. (1988). Cancer Res. 48, 1035-1041.

44. Strife, A. (1987) Blood 69:1508-1523.

45. ten Dijke, P. et al, (1990 A) International

Publication No. WO 90/14360, published November 29, 1990.

46. ten Dijke, P., et al., (1990 B) Annals of New York Academy. pp. 26-42.

47. Twardzik et al. (1989) J. Natl. Cancer Inst.

£1(15):1182.

48. Wong, et al (1975) Proc, Natl. Acad. Sci. USA 72:3167-3171.

49. Wrann, M. et al. (1987) EMBO J, 6: 1633-1636.

i’AOI'ER\Mr&lt;OW&gt;28.93.SPE-29/7/97