The invention provides a method of treating diabetes in a subject, comprising administering to the diabetic subject an immunotoxin, thereby reducing the subject's T-cell population, and administering to the subject pancreatic islet cells from a donor. The immune tolerance inducing treatment regimen, used optionally with adjunct immunosuppressive agents, prevents pancreatic islet cell rejection while maintaining long term islet cell function following xenogeneic and allogeneic pancreatic islet cell transplantation. Thus, the methods of the present invention provide a means for treating diabetes, wherein the need for exogenous insulin or immunosuppressive agents is decreased or eliminated. Also provided is a method of inhibiting a rejection response of a transplant recipient, comprising administering an immunotoxin during the peritransplant period, thereby transiently reducing the number of T-cell lymphocytes and promoting long-term survival of the transplant.

SUMMARY OF THE INVENTION

The invention provides a method of treating diabetes in a subject, comprising administering to the diabetic subject an immunotoxin, thereby reducing the subject's T-cell population, and administering to the subject pancreatic islet cells from a donor. The method is effective for treating Type I or Type II diabetes. More specifically, the invention provides a short course immune tolerance inducing treatment regimen utilizing an anti-CD3 immunotoxin that, optionally with adjunct immunosuppressive agents, prevents pancreatic islet cell rejection while maintaining long term islet cell function following pancreatic islet cell transplantation. The invention further provides sufficient immune tolerance to pancreatic islet transplantation so that xenogeneic transplants are not rejected. Thus, the methods of the present invention provide a means for treating diabetes, wherein the need for exogenous insulin or immunosuppressive agents is eliminated.

A method of inducing immune tolerance using an immunotoxin for xenogeneic or allogeneic pancreatic islet cell transplantation is provided. Also provided is a method of inhibiting a rejection response of a transplant recipient, comprising administering an immunotoxin during the peritransplant period, thereby transiently reducing the number of T-cell lymphocytes and promoting long-term survival of the transplant.

DETAILED DESCRIPTION OF THE INVENTION

The current invention provides a short course immune tolerance inducing treatment regimen utilizing an anti-CD3 immunotoxin that, optionally with adjunct immunosuppressive agents, prevents pancreatic islet cell rejection while maintaining long term islet cell function. A unique feature of the present immunotoxin immune tolerance induction is that the ratio of target cell toxicity (T cell) to non-target cell toxicity (islet cell) is extremely high. The adjunct immunosuppressive agents, which can also be used and which can cause islet cell toxicity, such as cyclosporine and corticosteroid derivatives, are required only for a brief duration due to the power of the primary immunotoxin promoting immune tolerance.

The invention provides a method of treating diabetes in a subject, comprising administering to the diabetic subject an immunotoxin, thereby reducing the subject's T-cell population, and administering to the subject pancreatic islet cells from a donor. Specifically, the method is effective for treating Type I or Type II diabetes. By "diabetes" is meant diabetes mellitus, a metabolic disease characterized by a deficiency or absence of insulin secretion or responsivity of peripheral tissues to insulin. As used throughout, "diabetes" includes Type I, Type II, Type III, and Type IV diabetes mellitus unless specified otherwise.

Unexpectedly, the immune tolerance induction regimen of the invention successfully reverses type II diabetes in monkeys. The reversal of the insulin resistant state in non-human primates suffering from type II diabetes implies that the peripheral insulin resistance seen in this disease is, in some way, mediated by faulty islet function, and this condition can be largely reversed by healthy grafted islets. The present invention can reduce or eliminate the need for insulin replacement in the subject with Type I diabetes as well. In the case of both Type I and Type II, non-fasting blood glucose levels will preferably be maintained below 160 mg/dl upon completion of the immune tolerance induction regimen.

By "pancreatic islet cells" is meant a composition comprising pancreatic islet cells. Preferably the pancreatic islet cells can be transplanted by injection of the cells into the portal vein; however, other cell, tissue, and organ transplantation paradigms well known in the art can be used. It is comtemplated that the immunotherapeutic function of the present immunotolerance tolerance induction regimen can be applied to transplantation of all or part of the pancreas as well as to the transplantation of pancreatic islet cells.

The "donor" can be a cadaver or a living donor. Furthermore, the donor can be of the same species as the subject being treated or a different species than the subject being treated. Thus, using the method of the invention, transplantation can be performed across primate species (i.e., xenogeneic transplantation or xenograft) and within the same primate line (i.e., allogeneic transplantation or allograft). It was not obvious until this invention that the highly sensitive pancreatic islet xenografts would maintain function in the presence of this immune tolerance inducing regimen. Nor was it obvious that this regimen would permit an islet transplantation crossing species lines.

The invention further provides an immune tolerance inducing regimen, wherein a population of the pancreatic islet cells to be transplanted are modified to decrease antigenicity prior to transplantation. Specifically, the donor cells can be altered, such as by genetically engineering the donor or donor cells, to reduce pancreatic islet cell antigenicity or to reduce the susceptibility of pancreatic islet cells to immune injury (R. Weiss, Nature 391: 327 28 (1998)).

The "subject" being treated can include individual humans, domesticated animals, livestock (e.g., cattle, horses, pigs, etc.), and pets (e.g., cats and dogs).

The invention further provides a method, wherein the immunotoxin transiently reduces the subject's T cells in the blood and lymph nodes by at least one log unit. Preferably the number of T cells in the blood and lymph nodes will be transiently decreased by 0.7, 0.8, 0.9, 1, 1.1, 1.2, 1.3, 1.4, 1.5, 1.6, 1.7, 1.8, 1.9, 2 log units, or any interval amount between 0.7 and 2 log units.

By "transiently reduces" is meant that T cells are reduced by 0.7 to 2 log units in the blood and lymph compartments for at least four days before starting to return to normal levels.

The present method of inducing immune tolerance or treating diabetes further comprises administering an immunosuppressive agent to the subject. The immunosuppressive agents can be administered beginning 24 to 0 hours prior to administration of the pancreatic islet cells to the recipient and continuing up to two weeks thereafter. Preferably, the immunosuppressive agents can be administered for at least 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14 days, or any interval time. The immunosuppressive agent is selected from the group consisting of cyclosporine, mycophenolate moefitil, methyl prednisolone, deoxyspergualin, other known immunosuppressive agents, and any combination thereof The immunotoxin regimen allows short-term immunosuppressive therapy and eliminates the need for long-term treatment with immunosuppressives, thereby avoiding the side effects associated with chronic immunosuppression.

The immunotoxin is administered beginning at 24 to 0 hours before administration of the pancreatic islet cells and continuing up to several days thereafter. Preferably, the immunotoxin is administered for 1, 2, or 3 days or any time in between. It is contemplated that immunotoxin administration of 4, 5, 6, or 7 days can be used if the production of antitoxin antibodies, which begins after approximately 5 days of administration, can be addressed. The immunotoxin can be administered in subjects beginning anytime after administration of the pancreatic islet cells. Thus, it is contemplated that a subject with a long term surviving transplant, who have not previously received immunotoxin treatment, could still benefit from immunotoxin administration, thereby avoiding the need for long-term treatment with immunosuppressives. It is further contemplated that a transplant recipient who begins to show signs of rejection may benefit from immunotoxin administration to reduce or eliminate the rejection process.

The invention further provides a method of inhibiting a rejection response of a recipient of a pancreatic islet transplant by inducing immune tolerance in the recipient, comprising administering an immunotoxin during the peritransplant period, thereby transiently reducing the number of T-cell lymphocytes and promoting long-term survival of the transplant. By "peritransplant period" is meant the period between 24 hours prior to and 2 weeks after transplantation.

Because of the close similarities between the immune functions that govern graft acceptance and rejection within primates and the similar functions of the pancreata within primates, this tolerance induction regimen is considered likely to succeed in humans for the treatment of diabetes via islet transplantation.

The immunotoxin can be a divalent anti-T cell immunotoxin, such as UCHT1-CRM9. More specifically, the divalent anti-T cell immunotoxin can be a single chain engineered fusion protein comprising an amino-terminus DT based toxin domain fused to a sFv domain (VL-linker-VH where linker is (Gly4Ser)3 separated by a second linker and fused to a second identical sFv domain. The second linker can be (Gly4Ser)3 or (Gly4Ser)5. Alternatively, the single chain engineered fusion protein can be monovalent providing that the linker and the VL and VH sequence are carefully chosen to provide an affinity lying within .+-.0.5 log unit of the parental antibody affinity.

The divalent disulfide linked immunotoxin can have identical components. Alternatively, the components can be non-identical with only one toxin moiety to minimize stearic hindrance of the antigen binding domains. The divalent anti-T cell immunotoxin can be a disulfide dimer of two monovalent single chain engineered fusion proteins that are dimerized via the hinge region of IgG1 or the .mu.CH2 domain of IgM. The dimer can be a homo dimer comprising two monovalent units of DT390-sFv-H-.gamma. CH3, disulfide dimerized by the single or double cysteine residues in H the hinge region. The dimer alternatively can be a heterodimer comprising one monovalent unit of DT390-sFv-H-.gamma.CH3, disulfide dimerized by the single or double cysteine residues in H to a monovalent unit of sFv-H-.gamma.CH3. Dimerization can be achieved in vivo by expression in a eukaryotic expression system modified for toxin resistance by an EF2 mutation. Examples are mutated CHO cells, Ss9 insect cells, or mutated yeasts. Alternatively, dimerization can be performed in vitro from monovalent species produced in prokaryote expression systems by the use of disulfide interchange reactions employing suitable disulfide oxidation systems such as dithiobisnitrobenzoic acid used to generate mixed disulfide intermediates.

The divalent anti-T cell immunotoxin can comprise a toxin moiety and a targeting moiety directed to the T cell CD3.di-elect cons. epitope. Even more specifically, the toxin moiety can be a diphtheria toxin binding site mutant. The immunotoxin can comprise a mutant toxin moiety (e.g., DT toxin or ETA toxin) linked to a single chain (sc) variable region antibody moiety (targeting moiety). Thus, the invention utilizes an immunotoxin having recombinantly produced antibody moiety linked (coupled) to a recombinantly produced toxin moiety and a fusion immunotoxin (where both toxin and antibody domains are produced from a recombinant construct). As the application provides the necessary information regarding the arrangement of toxin and antibody domains, and the sub regions within them, it will be recognized that any number or chemical coupling or recombinant DNA methods can be used to generate an immunotoxin. Thus, reference to a fusion toxin or a coupled toxin is not necessarily limiting.

The antibody moiety preferably routes by the anti-CD3 pathway. The immunotoxin can be monovalent, but divalent antibody moieties are presently preferred since they have been found to enhance cell killing by about 3 to 15 fold. The immunotoxin can be a fusion protein produced recombinantly. The immunotoxin can be made by chemical thioether linkage at unique sites of a recombinantly produced divalent antibody (targeting moiety) and a recombinantly produced mutant toxin moiety. The targeting moiety of the immunotoxin can comprise the human .mu.CH2, .gamma.CH3 and .mu.CH4 regions and VL and VH regions from murine Ig antibodies. These regions can be from the antibody UCHT1 so that the antibody moiety is scUCHT1, which is a single chain CD3.di-elect cons. antibody having human .mu.CH2, .gamma.CH3 and .mu.CH4 regions and mouse variable regions. These are believed to be the first instances of sc anti-CD3 antibodies. Numerous DT mutant toxin moieties are described herein, for example DT390 or extensions of DT390 out to DT530. Thus, as just one specific example the immunotoxin, the invention provides scUCHT1-DT390. Derivatives of this immunotoxin are designed and constructed as described herein.

The engineered divalent immunotoxin utilized in the present invention can be stabilized with respect to divalency and disulfide bond initiation by the interactive Ig domains of either human IgM .mu.CH2 or IgGI .gamma.CH3 or other similar Ig interactive domains.

In the present invention, the toxin moiety retains its toxic function and membrane translocation function to the cytosol in full amounts. The loss in binding function located in the receptor binding domain of the toxin protein diminishes systemic toxicity by reducing binding to non-target cells. Thus, the immunotoxin can be safely administered. The routing function normally supplied by the toxin binding function is supplied by the targeting antibody, for example, anti-CD3. The essential routing pathway is (1) localization to coated pits for endocytosis, (2) escape from lysosomal routing, and (3) return to the plasma membrane.

The mutant DT toxin moiety can be a truncated mutant, such as DT390, extensions of DT390 out to DT530, or other truncated mutants, as well as a full length toxin with point mutations or CRM9 (cloned in C. ulcerans), scUCHT1 fusion proteins with DTM1 and DT483. DT390 has been cloned and expressed in E. coli. The toxin domain can be CRM9 (DT535, S525F) plus a second attenuating mutation from the group: F530A, K516E, K516A, Y514A, V523A, N524A). The antibody moiety can be scUCHT1 or other anti-CD3 antibody having the routing and other characteristics described in detail herein. Thus, one example of an immunotoxin for use in the present methods is the fusion-protein immunotoxin DT390sSvUCHT1. In principal, described immunotoxins can be used in the methods of the invention.

The immunotoxin or components thereof can be expressed in E. coli BL21DE3 cytosol using TrxB.sup.-strains at 15 to 25.degree. C. Alternatively, the immunotoxin or components thereof can be expressed in eukaryotic cell lines (such as CHO), insect cell lines (such as Ss9), and yeast cell lines (such as Pichia pastoris) provided that the toxin glycosylation sites at residues 16 and 235 are eliminated and the cells have been mutated to resist ADP-ribosylation catalyzed by toxin, by a Gly to Arg substitution 2 residues to the carboxyl side of the modified amino acid diphthamide. In the case of insect cells this mutant EF-2 can be supplied in the same baculovirus vector supplying the immunotoxin gene since two late promoters are available in baculovirus.

The recombinant immunotoxins can be produced from recombinant sc divalent antibody or recombinant dicystronic divalent antibody and recombinant mutant toxins each containing a single unpaired cysteine residue. An advantage of this method is that the toxins are easily produced and properly folded by their native bacteria while the antibodies are better produced and folded in eukaryote cells. In addition, this addresses differences in coding preferences between eukaryotes and prokaryotes which can be troublesome with some immunotoxin fusion proteins.

The general principles for producing the present divalent recombinant anti-T cell immunotoxins are:

1. The disulfide bond bridging the two monovalent chains is chosen from a natural Ig domain, for example from .mu.CH2 (C337 of residues 228 340 or the .gamma.IgG hinge region, C226 or C229 or both of residues 216 238 [with C220P]).

2. Sufficient non-covalent interaction between the monovalent chains is supplied by including domains having high affinity interactions and close crystallographic or solution contacts, such as .mu.CH2, .mu.CH4 (residues 447 576) or .gamma.CH3 (residues 376 446). These non-covalent interactions facilitate proper folding for formation of the interchain disulfide bond.

3. For fusion immunotoxins the orientation of the antibody to the toxin is chosen so that the catalytic domain of the toxin moiety becomes a free entity when it undergoes proteolysis at its natural processing site under reducing conditions. Thus, in the ETA based IT, the toxin moiety is at the carboxy terminus and, in DT based fusion IT, the DT based toxin moiety is at the amino terminus of the fusion protein.

4. For chemically coupled immunotoxins, a single cysteine is inserted within the toxin binding domain. The antibody is engineered to have only a single free cysteine per chain which projects into the solvent away from interchain contacts such as .mu.CH3 414, .mu.CH4 575 or the addition to .gamma.CH3 at C447. Crystal structure indicates this region is highly solvent accessible. Excess free cysteines are converted to alanine. Alternatively, a C terminal cysteine can be added to .gamma.CH3 directly following a histidine tag for purification, .gamma.CH3 (His)6Cys.

5. Toxins are mutated in their binding domain by point mutations, insertions or deletions, have at least a 1000 fold reduction in binding activity over wild type, and are free of translocation defects.

6. Toxin binding site mutants, if not capable of proteolytic processing at neutral pH, are modified in the processing region to achieve this result.

A binding site mutant (CRM9) of full length diphtheria toxin residues 1 535 using the numbering system described by Kaczovek et al. (56) S525F (57) can be further modified for chemical coupling by changing a residue in the binding domain (residues 379 535) to cysteine. Presently preferred residues are those with exposed solvent areas greater than 38%. These residues are K516, V518, D519, H520, T521, V523, K526, F530, E532, K534 and S535 (57). Of these K516 and F530 are presently preferred since they are likely to block any residual binding activity (57). However, maximal coupling of the new cysteine residue will be enhanced by the highest exposed solvent surface and proximity to a positively charged residue (which has the effect of lowering cysteine--SH pKa). These residues are at D519 and S535 so that these are presently preferred from the above list of possibilities.

A double mutant of DT containing the S525F mutation of CRM9 plus an additional replacement within the 514 525 exposed binding site loop to introduce a cysteine coupling site for example T521C can be produced in Corynebacterium ulcerans preceded by the CRM9 promoter and signal sequence. The double mutant is made in Corynebacterium ulcerans by a recombination event between the plasmid producing CRM9-antibody fusion protein and PCR generated mutant DNA with a stop codon at 526 (gapped plasmid mutagenesis). Alternatively, double attenuating mutants of diphtheria toxin can be produced in E. coli BL21DE3 TrxB.sup.- by PCR mutagenesis without the use of a signal sequence. These CRM9-Cs can be used to form specific thioether mutant toxin divalent antibody constructs by adding excess bismaleimidohexane to CRM9-Cs and coupling to single chain divalent antibody containing a free cysteine at either the end of the .mu.CH4 domain or the .gamma.CH3 domain (see Ser. No. 08/739,703, hereby incorporated by reference).

These and other mutations are accomplished by gapped plasmid PCR mutagenesis (58) using the newly designed E. coli/C. ulcerans shuttle vector yCE96 containing either the double mutant DT S508F S525F or a CRM9 COOH terminus fusion protein construct having reduced toxicity due to the COOH terminal added protein domain (59).

The mutated toxins are produced and purified analogously to the parent toxin except that low levels of reducing agent (equivalent to 2 mM betamercaptoethanol) are included in the purification to protect the unpaired introduced --SH group. Thioether chemical coupling is achieved to a single unpaired cysteine within the divalent antibody construct at either residue 414 in domain .gamma.CH-3 or residue 575 in domain .mu.CH4 when this domain is included. In this case domain .gamma.CH-3 is mutated C414A to provide only a single coupling site. An advantage of including .mu.CH4 is enhanced stability of the divalent antibody. A disadvantage is that the extra domain increases size and thereby reduces the secretion efficiency during antibody production. The advantage of terminating with the .gamma.CH3 domain is that, in another variant, a His6 purification tag can be added at either the .mu.CH2 COOH or .gamma.CH3 COOH terminus to facilitate antibody purification. Another variant is to use the .gamma. hinge region to form the interchain disulfide and to couple through a .gamma.CH3 or .mu.CH4. This variant has the advantage of being smaller in size and places the toxin moiety closer to the CD3 epitope binding domains, which could increase toxin membrane translocation efficiency. A His tag can be included at the carboxy terminus as a purification aid. SH-CRM9 is concentrated to 10 mg/ml in PBS pH 8.5 and reacted with a 15 fold molar excess of bismaleimidohexane (BW (Pierce, Rockford, Ill.). Excess BMH is removed by passing over a small G25F column (Pharmacia, Piscataway, N.J.). The maleimide derived toxin at about 5 mg/ml is now added to scUCHT1 divalent antibody at 10 mg/ml at room temperature. After 1 hr the conjugate is separated from non-reactive starting products by size exclusion HPLC on a 2 inch by 10 inch MODcol column packed with Zorbax (DuPont) GF250 6 micron resin (for large scale production). Derivatives of ETA60EF61cys161 are also coupled to scUCHT1 divalent antibody by the same method.

Divalent anti-T cell fusion immunotoxins based on DT can be utilized in the invention, wherein the toxin domain (also referred to herein as "toxin moiety" or "tox") is either full length mutant S525F (CRM9) or truncated at 390 or 486 (collectively Tox) and the sequence of domains from the amino terminus from left to right can be selected from among the following and may include C terminal or amino terminal His purification tags: VL and VH are the variable light and heavy domains of the anti-CD3 antibody UCHT1 or other anti-CD3 antibody. H is the human IgG1 hinge.

Single chain divalent fusion protein: Tox, VL, L, VH, L, VL, L, VH;

Single chain univalent fusion protein homodimerized via .mu.CH2 337 Cys: (Tox, VL, L, VH, .mu.CH2)2;

Single chain univalent fusion protein homodimerized via H 226/229 Cys: (Tox, VL, L, VH, H, .gamma.CH3)2;

Single chain univalent fusion protein heterodimerized via .mu.CH2 337 Cys: (Tox, VL, L, VH, .mu.CH2 VL, L, VH, .mu.CH2);

Single chain univalent fusion protein heterodimerized via H 226/229 Cys: (Tox, VL, L, VH, H, .gamma.CH3 VL, L, VH, H, .gamma.CH3);

sFv-SH fusion protein homodimerized via H 226/229 Cys chemically linked via a bis maleimide (R) to a SH derivatized CRM9 or a CRM9 containing an engineered C terminal cysteine (Tox): VL, L, VH, H, .gamma.CH3, His6, Cys-SH-R-SH-Tox.

Other types of protein toxin moieties can be utilized in anti-T cell immunotoxins for the induction of tolerance. All that is required is that a 1-2 log kill of T cells within the blood and lymph node compartments can be achieved without undue systemic toxicity. This in turn requires that the routing epitope routes in parallel with the toxin intoxication pathway and that binding site mutants are available or that toxins truncated in their binding domain are available that reduce toxin binding by 1000 fold compared to wild type toxins without compromising toxin translocation efficiency (see U.S. Pat. No. 5,167,956 issued Dec. 1, 1992). In addition when using targeting via antibodies, divalent antibodies are generally required under in vivo conditions to achieve sufficient cell killing due to the 3 to 15 fold lower affinity of monovalent antibodies. However, the method of linking the toxin to the divalent antibody either as a single chain fusion protein or through specific engineered coupling sites must not interfere with translocation efficiency. This could occur due to the larger size of many divalent antibodies compared to monovalent scFv antibodies unless care is taken so that the catalytic domain of the toxin can achieve unencumbered translocation. This is achieved for DT based immunotoxins using DT based binding site mutants where the fusion protein antibody moiety is contiguous with the COOH terminus of the toxin binding chain as described above. This allows the catalytic a chain to translocate as soon as the disulfide loop spanning the Arg/Ser proteolytic processing site residues 193/194 is reduced. Most targeted cells are capable of performing this processing event, and when chemically coupled CRM9 is used the processing is performed by trypsin prior to coupling.

If the toxin moiety is based on full length diphtheria toxin, it can include the following mutations:

S525F, K530C

S525F, K516C

S525F, D519C

S525F, S535C.

The antibody-toxin constructs utilized in the invention can be expected to be effective as immunotoxins because the relevant parameters are known. The following discussion of parameters is relevant to the use of the immunotoxin in tolerance induction. The relevant binding constants, number of receptors and translocation rates for humans have been determined and used. Binding values for anti-CD3-CRM9 for targeted and non-targeted cells in vitro and rates of translocation for the anti-CD3-CRM9 conjugate to targeted and non-targeted cells in vitro are described (Greenfield et is al. (1987) Science 238:536; Johnson et al. (1988) J. Biol. Chem. 263:1295; Johnson et al. (1989) J. Neurosurg. 70:240; and Neville et al. (1989) J. Biol. Chem. 264:14653). The rate limiting translocation rate to targeted cells in vitro is shown as follows: an anti-CD3 -CRM9 conjugate at 10.sup.-11 M is translocated to about 75% of the target cells present as measured by inhibition of protein synthesis in about 75% of cells with 20 hours. Inhibition of protein synthesis is complete in cells into which the conjugate translocates.

Parameters determined in in vivo studies in nude mice include the following: Tumor burden is described in Example 1 as a constant mass equal to 0.1% of body weight; the receptor number and variation of receptor number are described in Example 3; "favorable therapeutic margin" is defined as an in vivo target cell 3 log kill at 0.5 MLD (minimum lethal dose)-comparison of efficacy with an established treatment of 0.5 MLD immunotoxin equivalent (group 1) to a radiation dose of 500 600 cGy (groups 8 and 9).

The parameters determined in vitro allowed the prediction of success in the in vivo nude mouse study. The prediction of in vivo success was verified by the data in Examples 3 4. Using the target cell number from the mouse study as being equivalent to the local T cell burden in a monkey or man successful T cell ablation and immunosuppression in monkeys could be predicted. This prediction has been verified by the monkey data in Example 5. Using the same parameters, a scientist skilled in this field can make a prediction of success in humans with confidence, because these parameters have been previously shown to have predictive success.

Most human sera contain anti-DT neutralizing antibodies from childhood immunization. To compensate for this the therapeutic dose of anti-CD3-CRM9 can be appropriately raised without affecting the therapeutic margin. Alternatively, a non-toxic DT mutants reactive with neutralizing antisera (e.g., CRM197) that can be administered in conjunction with the immunotoxin. See U.S. Pat. No. 5,725,857, the content of which is incorporated herein by reference.

One embodiment to the invention provides a method of inhibiting a rejection response by inducing immune tolerance in a recipient to a foreign donor pancreatic islet cell by exposing the recipient to an immunotoxin so as to reduce the recipients's peripheral blood T-cell lymphocyte population by at least 80%, and preferably 95% or higher, wherein the immunotoxin is an anti-CD3 antibody linked to a diphtheria protein toxin, and wherein the protein has a binding site mutation. The term "donor cell" refers to a donor pancreatic islet cell or cells, as distinguished from donor lymphocytes or donor bone marrow. When the donor pancreatic islet cells are transplanted into the recipient, a rejection response by the recipient to the donor cells is inhibited and the recipient is tolerized to the donor organ or cell. Alternatively, a non-toxic DT mutant such as DTM2 or CRM197 can first be administered followed by the immunotoxin. This method can use any of the immunotoxins (e.g., anti-CD3-CRM9, scUCH1-DT390, etc.) or non-toxic-DT mutants described herein with the dosages and modes of administration as described herein or otherwise determined by the practitioner.

The present tolerance induction method can also include administering an immunosuppressant compound before, at the same time, or after the immunotoxin exposure step. The immunosuppressant compound can be cyclosporine (e.g., cyclosporine A) or other cyclophylins, mycophenolate moefitil (Roche), methyl prednisone, deoxyspergualin (Bristol Myers), FK506, other known immunosuppressants, or any combination thereof. It will be appreciated that certain of these immunosuppressants have major effects on cytokine release occurring in the peritransplant period that may aid in the induction of the tolerant state. The method of inducing immune tolerance can further comprise administering donor bone marrow at the same time or after the exposure step.

Any one, two, or more of these adjunct therapies can be used together in the present tolerance induction method. Thus, the invention includes at least six methods of inducing tolerance using immunotoxin (IT): (1) tolerance induction by administering IT alone; (2) tolerance induction by administering IT plus other drugs that alter immune function such as high dose corticosteroids; (3) tolerance induction by administering IT plus immunosuppressant drugs, such as mycophenolate mofetil and/or deoxyspergualin; (4) tolerance induction by administering IT plus other drugs that alter immune function, plus immunosuppressant drugs; (5) tolerance induction by administering IT and bone marrow; and (6) tolerance induction by administering IT plus bone marrow, plus immunosuppressant drugs. The adjunct therapy can be administered before, at the same time or after the administration of immunotoxin. Different adjunct therapies can be administered to the recipient at different times or at the same time in relation to the transplant event or the administration of immunotoxin, as further described below.

Because the immunosuppressant can be administered before the immunotoxin and/or other treatments, the present method can be used with a patient that has undergone an organ transplant and is on an immunosuppressant regimen. This presents a significant opportunity to reduce or eliminate traditional immunosuppressant therapy and its well documented negative side-effects. Also, as described below, treatment with immunosuppressants prior to transplantation could be particularly useful in cadaveric and xenogeneic transplants. In such a setting of pre-transplant treatment with immunosuppressant, the administration of immunotoxin can be delayed for up to seven or more days post-transplantation.

An example of a schedule of immunotoxin and immunosuppressant administration for patients receiving organ transplants is as follows: day -24 to -0 hours: begin immunosuppressant treatment; day 0: perform transplant; day 0: immediately following transplant administer 1st immunotoxin dose; day 1: 2nd immunotoxin dose; day 2: 3rd and final immunotoxin dose. Immunosuppressant treatment may end at day 3 or extend to day 14. Immunosuppressant treatment is also effective if begun at the time of transplantation, and can continue for up to two weeks after transplantation.

The presently preferred doses of the immunotoxin are those sufficient to deplete peripheral blood T-cell levels to 80%, preferably 90% (or especially preferably 95% or higher) of preinjection levels. This should require mg/kg levels for humans similar to those for monkeys (e.g., 0.05 mg/kg to 0.3 mg/kg body weight), which toxicity studies indicate should be well tolerated by humans. Thus, the immunotoxin can be administered to safely reduce the recipients T cell population.

Claim 1 of 8 Claims

1. A method of transplanting cadaveric donor pancreatic islet cells to a subject in need thereof, comprising (a) administering to the subject a divalent anti-T cell diphtheria toxin binding site mutant immunotoxin directed at the CD3 epitope during the peritransplant period, thereby reducing the subject's T-cell population; (b) administering deoxyspergualin to the subject; and (c) administering to the subject pancreatic islet cells from a cadaveric donor.
 

____________________________________________
If you want to learn more about this patent, please go directly to the U.S. Patent and Trademark Office Web site to access the full patent.