INVESTIGATOR’S BROCHURE

BUTHIONINE SULFOXIMINE NSC 326231 IND # 69,112 Initial Version for IND 32,981 June 1994 (Revised June 1996) Division of Cancer Treatment National Cancer Institute Diagnosis and Centers Bethesda, Maryland

Revised December 2003 Version 2.0 Developmental Therapeutics Program USC-CHLA Institute for Pediatric Clinical Research Childrens Hospital Los Angeles Los Angeles, CA

This document contains both published and unpublished data describing the results of preclinical studies with this agent. The unpublished data are preliminary and should not be quoted or used for any purpose other than as guidance in developing clinical trials.

Table of Contents Page

Summary

3

Introduction

5

Pharmaceutical Information

10

Mechanism of Action

12

Preclinical Activity

18

GSH Depletion and Effect on L-PAM Cytotoxicity

18

Tissue Selectivity of BSO-Mediated GSH Depletion

21

Tumor Cell Selectivity

23

Effect of BSO and L-PAM on Survival

27

BSO Sensitization of Agents Other Than L-PAM

31

Preclinical Toxicology and Pharmacology

36

Experience with BSO in Humans

50

Summary of Data and Guidance for the Investigator

56

References

58

Appendix

Attached

2

SUMMARY Glutathione (GSH) is a ubiquitous intracellular tri-peptide that protects cells from oxidative stress and plays an important role in detoxifying alkylating agents such as melphalan (L-PAM) via dechlorination, quenching of DNA adducts, and conjugation reactions catalyzed by glutathione S-transferase.1-3

L-S,R-buthionine sulfoximine (BSO), a specific inhibitor of γ-

glutamyl cysteine synthetase (the rate-limiting enzyme in GSH synthesis), depletes GSH in vitro and in tumors in vivo, and BSO has been shown in preclinical

studies, to reverse alkylator

resistance.2; 4-7 BSO, as a single agent, is highly cytotoxic for neuroblastoma cell lines in vitro and results in apoptosis due to increased generation of reactive oxygen species (ROS).8 ;

9

ROS that

are elaborated during catecholamine synthesis are the hypothesized etiology of this unique single agent BSO activity which has not been reported for other pediatric tumors such as rhabdomyosarcoma, Ewing’s Family Tumors, or retinoblastoma.

The in vitro combination of

BSO with clinically achievable (non-myeloablative) concentrations of L-PAM was highly synergistic for neuroblastoma cell lines established at diagnosis or following non-myeloablative therapy and led to the a recently completed pediatric pilot study (see appendix).10;

11

Preclinical

toxicology studies in animals showed no major toxicity at doses of BSO used in human clinical trials.12; 13 Clinically, BSO has been given to adults, alone and in combination with L-PAM.

The

most typically used dosing schedule has been a loading dose of 3 g/m2 bolus over 30 minutes i.v. followed by a 72 hour continuous infusion of 0.75 g/m2 /hr and L-PAM (15 mg/m2 bolus at hour

3

48 of BSO infusion).

A pilot study in pediatrics tested the same schedule and also tested

increasing the c.i.v. for BSO to 1.0 g/m2 /hr.10; 11 The reported systemic toxicities of BSO and L-PAM in adults were unremarkable with reversible myelosuppression being the most common.14-16

Documented responses have occurred

in patients with melanoma (complete response 18 months), ovarian cancer, breast cancer, and small cell carcinoma of the lung.14-16 A pilot study of BSO combined with 15 mg/m2 of L-PAM in children with recurrent/refractory high-risk neuroblastoma achieved a 29% response rate (duration 1-3 months), with 7 partial (6 of 7 had previous myeloablative therapy) and 2 minor response (Preprint in Appendix).10;

11

Toxicity in the pediatric study was similar to that in adults,

with most patients showing grade 3-4 leukopenia and thrombocytopenia.

The degree of

hematopoietic toxicity seen with BSO + L-PAM appears to be increased over what would be expected for L-PAM alone at the same dose. In contrast to the adult studies, there were 2 deaths due to CNS toxicity in the first pediatric study of BSO + L-PAM.

Both patients having a toxic

death presented with acute tubular necrosis; one had a large, intracranial mass.

No other

significant toxicities were seen. In addition to the possibility of the CNS mass pre-disposing to toxicity from BSO + L-PAM, there is the possibility that a drug interaction between BSO and cephalosporin antibiotics could lead to BSO toxicity (preprint in appendix).11

Thus, patients

receiving BSO should not be given cephalosporin antibiotics, or any agent known to deplete GSH, such as acetaminophen.

The t½ of BSO is ~ 2.3 hours and BSO may decrease the

clearance of L-PAM, potentially increasing the L-PAM AUC.

Depletion of GSH by BSO in

blood mononuclear cells occurred to 30-40% of baseline in adults and ~46% in children, and appeared to return to normal within 4 days of stopping the BSO.

4

INTRODUCTION DL-Buthionine-S, R-sulfoximine (BSO) is an analog of methionine sulfoximine (MSO), which was identified as the toxic agent in agonized grain. synthetase and

MSO inhibits both glutamine

γ-glutamylcysteine synthetase (γ-GCS), enzymes which participate in the

synthesis of glutathione (γ-L-glutamyl-L-cysteinylglycine, GSH).

GSH is a ubiquitous tripeptide

found in mammalian cells and in other animal and plant cells, as well as in prokaryotic cells. GSH is the major intracellular nonprotein sulfhydryl and is involved in a variety of metabolic and physiologic functions, such as catalysis of enzymatic reactions, amino acid transport, detoxification of xenobiotics, and normal cellular reductive processes.

GSH exists in millimolar

concentrations (0.5-10 mM) under normal steady-state conditions in the cell, but can also exist in interchanging states as glutathione disulfide (GSSG), as part of mixed disulfides, or as a thiol ester. Biosynthesis of GSH proceeds through two successive reactions which require ATP.

Although MSO was effective in depleting intracellular thiols when tested in vitro using isolated synthetases, it did not possess the specificity required for in vivo studies and was also inherently toxic.

Consequently, analogs of MSO, such as BSO and prothionine sulfoximine

(PSO), were synthesized and evaluated for specificity of GSH inhibition and activity both in vitro and in vivo. Among the sulfoximine derivatives examined, BSO was the most specific and least toxic.

BSO was noted to selectively bind to the active site of γ-GCS, the enzyme that

catalyzes the first of two steps in the synthesis of GSH. The synthesis of GSH and the structure and site of action for BSO is shown in Figure 1.

5

BSO Inhibits Synthesis of Glutathione γ-glutamylcysteine synthase

GSH synthase

cysteine γ-glu-cys glutamine

CH3 CH2 H C CH2 H

BSO

NH S O (CH2)2 H C NH3 COO-

γ-glu-cys-gly

Glutathione glycine (GSH)

Figure 1 Once γ-GCS is inactivated by BSO, de novo GSH synthesis is inhibited. This results in depletion of intracellular GSH, especially under conditions that promote enhanced GSH utilization.

BSO has demonstrated tissue selectivity with respect to GSH depletion, and this

observation may be of clinical importance.

In an in vivo study in mice, GSH was maximally

depleted (70-89%) in certain tissues such as the liver and kidney and minimally depleted (40%) in the bone marrow. This reduced effect on bone marrow is thought to be responsible for BSO’s lack of myelosuppression. Intracellular levels of GSH have been shown to modulate the toxicity of various antineoplastic agents, such as cytotoxics and radiation therapy. action of GSH, in this regard, is not fully understood.

However, the mechanism of

The following mechanisms have been

proposed: 1) GSH may detoxify antineoplastics such as melphalan (L-PAM) via conjugation; 2) GSH may promote repair cellular injury resulting from the action of antineoplastics or radiation, through its role in oxidative-reductive or nucleophilic reactions; formation of active species which cause cellular injury;

4)

3) GSH may mediate the GSH is known to decrease

intracellular reactive oxygen species and the latter promote the toxic effects of radiation and some antineoplastic drugs or, 5) GSH may interfere with delivery or uptake of antineoplastic agents through its role in amino acid transport.

6

GSH and drug resistance.

The hypothesis that thiols may be involved in cellular

response to alkylating agents was explored. In one study, an inverse relationship was observed between the presence of protein and nonprotein sulfhydryls and the cytotoxicity of the alkylating agent, merophan. This prompted further studies to determine whether the major nonprotein thiol (GSH) plays a role in the modulation of the cytotoxicity of melphalan (L-PAM), an alkylating agent widely used in clinical oncology.

Results of these studies suggested that tumor cells

resistant to L-PAM contained higher levels of GSH than those sensitive to L-PAM. Data from these studies also demonstrated that nutritional deprivation of L-cysteine could reduce intracellular GSH to an equivalent extent in both L1210 murine leukemia cells sensitive to LPAM and L1210R cells resistant to L-PAM, which resulted in re-sensitization of L1210R cells to L-PAM to the same extent as L1210 cells grown under normal conditions. Some reports have indicated that the mechanism of cellular resistance to L-PAM is related to decreased L-PAM uptake.

However, other reports have shown no difference in L-

PAM uptake between L1210R or human ovarian cancer cells (1847me), both resistant L-PAM, and their respective L-PAM sensitive parent lines. Both resistant lines had approximately a 2fold higher intrinsic level of GSH.

Data from these studies indicated that L-PAM was more

readily converted to its noncytotoxic dechlorinated dihydroxy metabolite in the resistant than in the sensitive cells. Furthermore, the interaction between L-PAM and GSH was postulated to be mediated by GSH-S-transferase which normally is involved in the dehalogenation of several electrophiles. However, it is possible that this interaction may vary with cell type. Despite the role of extracellular L-cysteine as a modulator of intracellular GSH, nutritional deprivation of cysteine ws not considered a feasible therapeutic method to reduce GSH levels in vivo.

Consequently, BSO was explored as a thiol-depleting agent. Results

indicated the GSH could be effectively depleted in vitro in L1210R cells at a minimally toxic dose of BSO (50 µM), but that long exposure time (24 hours) was required for maximum depletion. These findings were confirmed in an in vitro study in which BSO effectively reduced GSH levels in both L-PAM resistant (1847me) and L-PAM-sensitive (A1847) human ovarian cancer cells.

Similar results were reported in another study of human ovarian cancer cells

(NIH:OVCAR-3), which were derived from patients clinically resistant to alkylataing agents.

7

The levels of GSH in the OVCAR-3 cells were intrinsically 4-fold higher than those in cells from untreated patients.

Treatments with a nontoxic dose (50 µM) of BSO reduced intracellular GSH

levels by 93.4% and decreased the inhibitory concentration (IC)50 of L-PAM by 73%, resulting in a dose modification factor (DMF) of 3.6. Despite the effectiveness of BSO in reducing intracellular GSH and increasing the cytotoxicity of L-PAM in vitro, two earlier in vivo studies did not indicate a significant increase in survival in mice bearing L1210R treated with BSO and L-PAM when compared to mice treated with L-PAM alone.

However, in a recent study, a significant increase in survival was

observed in mice bearing clinically resistant OVCAR-3 cells when the animals were treated with L-BSO and L-PAM.

The median survival time (MST) for the untreated mice was 46 days,

which increased to 68 days when the mice received L-PAM alone, and further increased to ≤ 125 days when L-BSO was administered prior to L-PAM. Although BSO is being evaluated only with L-PAM in the initial clinical trials, BSO has been shown in preclinical studies to increase the cytotoxicity of other cytotoxics, such as carmustine (BCNU), cisplatin (CDDP), doxorubicin (DOX), and cyclophosphamide (CTX), and radiosensitizers, such as SR-2508, misonidazole (MISO), and tirapazamine. BSO has also been shown to increase radiation response, although it is unclear as to the conditions (normoxic or hypoxic) required.

Treatment with BSO has shown promising activity in thermo sensitization

and inhibition of thermo tolerance. BSO has also been shown to enhance the anti-tumor activity anti-neoplastics acting via a variety of mechanisms, such as the retinoid fenretinide, dexamethasone, and arsenic trioxide in leukemia cell lines, and diethylstilbesterol in a prostate cancer model.

Although modulation of cytotoxicity by BSO may present a promising

therapeutic approach, the tissue selectivity of BSO-mediated GSH depletion may be a limiting factor in developing combination regimens which include BSO. that BSO increases the renal toxicity of MeCCNU.

Some reports have indicated

Consequently, for maximum therapeutic

benefit, BSO may be best combined with agents whose principal toxicity is myelosuppression. Preclinical toxicology.

Studies were conducted to determine the toxicity and

pharmacokinetic behavior of BSO in mice and dogs and to explore the effect of BSO on L-PAM –induced toxicity in mice.

For the toxicity studies of BSO alone, CD2F1 mice were given

8

multiple oral (PO) doses of BSO 100-800 mg/kg/dose or multiple intravenous (IV) doses of 4001500 mg/kg/dose.

Beagle dogs received a single IV dose of BSO 120 or 3200 mg/kg/dose,

multiple IV doses of 50,120, and 240 mg/kg/dose (every 12 hours x 6 doses, or multiple PO doses of 100-800 mg/kg/dose). BSO was toxic to the gastrointestinal, urinary, and central nervous systems of beagle dogs.

Possible hepatotoxicity was indicated by clinical pathology, but no treatment-related

microscopic lesions were found in the liver. BSO was lethal to dogs.

At the highest single PO or IV doses evaluated,

BSO produced severe GSH depletion in the dogs; however, a

relationship could not be established between the magnitude of GSH depletion and BSO-induced toxicity. With the multiple IV dose schedule, a slight decrease in weight was observed in one dog at the 240 mg/kg/dose level. Minor hematologic effects were noted among dogs receiving the 120 and 2340 mg/kg/dose schedule. Toxicity was reversible and was not sex-related in dogs. In contrast, mice tolerated large multiple PO or IV doses of BSO up to 800 mg/kg/dose without apparent toxicity, except for slight weight loss, and despite marked depletion of GSH levels. Transient depression in white blood cell count was observed in female mice at the highest IV dose level (1600 mg/kg/dose), but no bone marrow lesions were observed at necropsy. Pretreatment with BSO up to 800 mg/kg/dose PO or IV did not significantly alter the toxicity of L-PAM 5 mg/kg IV in mice. At high single IV doses (1600 mg/kg/dose), BSO potentitated LPAM-induced renal, bone marrow, and lymphoid toxicity in mice.

However, this potentiation

did not appear to be related to magnitude of GSH depletion. Preclinical pharmacology.

Elimination of drug from the plasma appeared to be

biexponential in either dogs or mice given BSO 2400 mg/m2 PO or IV. Although the initial and terminal half-lives (T ½) were similar in both species, the area under the plasma concentrations vs. time curve (AUC) in the initial phase accounted for 94% of the total AUC in mice and 20% of the total AUC in dogs. approximately 2% in mice.

Based on plasma drug levels, oral bio-availability of BSO was Since GSH depletion was comparable following either p.o. or i.v.

BSO, however, it is likely that the actual oral bioavailability is higher than that observed, possibly secondary to rapid removal of BSO from the plasma into tissue compartments.

9

INVESTIGATIONAL DRUG – PHARMACEUTICAL DATA BSO INJECTION (NSC-326231)

The following information applies to the investigational dosage form of BSO injection. The Division of Cancer Treatment, Diagnosis, and Centers, National Cancer Institute provided the product below to the current IND sponsor, The Developmental Therapeutics Program USCCHLA Institute for Pediatric Clinical Research, for clinical trials.

Chemical Name: Butanoic acid, 2-amino-4 (S-butylsulfonimidoyl), 2S CH3 CH2 H C CH2 H

NH S O (CH 2)2 H C NH3 COO-

Other Names: buthionine sulfoximine

Molecular Formula: C8 H18 N2 O3 S

M.S.: 222.3

How Supplied: Sterile injection, 100 mg/ml, 50 ml/vial:

The product is prepared as an aqueous solution with

sodium hydroxide for adjustment of pH to 7 to 8.

Storage: The intact vials should be stored at controlled room temperature (15 to 30° C)

10

Stability:

Shelf life surveillance of the intact vials is ongoing.

In aqueous solution, BSO is

most stable at neutral and basic pH values. At pH values of 6 to 9.9, little or no decomposition occurs in 48 hours at 90ºC. Increasing rates of decomposition occur at lower pH values. At a concentration of 0.1 mg/ml in 5% Dextrose Injection, USP, and 0.9% Sodium Chloride Injection, USP, in both glass bottles and PVC plastic bags, no decomposition or loss of drug due to sorption has been noted over a 14 day period.

Route of Administration: Intravenous

Distribution: The product has been supplied for authorized investigational studies to the USCCHLA Insittute for Pediatric Clinical Research by the Division of Cancer Treatment, Diagnosis, and Centers, National Cancer Institute, Bethesda, Maryland 20892, USA, via the Developmental Therapeutics Program, NCI.

Drug supplies for studies conducted under the current IND will be provided by:

Developmental Therapeutics Program USC-CHLA Institute for Pediatric Clinical Research Childrens Hospital Los Angeles 4650 Sunset Blvd Los Angeles, CA 90027

11

MECHANISM OF ACTION

The mechanism through which BSO modulates the activity of cytotoxic agents is related to two distinct actions; 1) the inhibition of GSH synthesis; and 2) the affect of intracellular GSH depletion on the activity of the antineoplastic agent.17-19 Inhibition of GSH synthesis by BSO is related to the BSO-mediated activation of ?-GCS, in the presence of ATP, catalyzes the formation of ?-glutamylcysteine (g-GC) from glutamate and cysteine; second, GSH synthetase, also in the presence of ATP, catalyzes the formation of GSH from ?GC and glycine.

?-Glutamyl phosphate is the enzyme-bound intermediate formed

from the action of ?-GCS on glutamate.20 BSO was anticipated to be a potent inhibitor of GSH synthesis due to its close structural resemblance to ?-glutamyl phosophate-a-amniobutyrate, which binds to ?-GCS,21 and its inhibition of de novo synthesis of GSH. Griffith22;

23

first attempted to elucidate the mechanism of inhibition of ?-GCS by BSO.

Their results were consistent with the proposed mechanism for PSO that the S-alkyl moiety of PSO binds at the site of ?-GCS to which the acceptor amino acid, cysteine or a-aminobutyrate, would normally bind.

BSO is thought to initially bind to the enzyme and then undergo

phosphorylation in the presence of ATP to the ?-GCS-bound BSO-phosphate, which is the actual ?-GCS inhibitor.21 Compared to the other sulfoximine derivatives examined, such as MSO and PSO, BSO was the most potent inhibitor of ?-GCS activity. BSO-induced inhibition of ?-GCS was at least 100-fold greater than that of MSO and approximately 20-fold greater than that of PSO. Complete inhibition of ?-GCS was achieved after a 10-minute incubation with 0.02 mM BSO. In contrast, complete inhibition of ?-GCS activity was not achieved after 10 minutes with 2 mM MSO. This high specificity of BSO for ?-GCS may account for its low toxicity relative to other thiol-depletors, such as MSO and PSO.23

12

The mechanism of ?-GCS inhibition by BSO was further examined by Griffith.24 In this study, it was established that ?-GCS, in the presence of MgATP, catalyzes the phosphorylation of BSO to form the BSO-phosphate adduct. The inhibition of ?-GCS by the enantiomers, L-BSO and D-BSO, was also examined.

At a concentration of 10 mM, D-BSO did not significantly

inhibit ?-GCS, whereas 20µM L-BSO resulted in complete inhibition of ?-GCS activity after only 10 minutes.

The rate of ?-GCS inactivation produced by incubation with 20 µM L-BSO

was equivalent to that produced by 40 µM DL-BSO (a mixture of the D and L enantiomers). Furthermore, unlike the reversibility of inactivation previously demonstrated with MSO,22 a 200fold dilution of BSO-bound γ-GCS did not result in enzyme reactivation. The inhibition kinetics of γ-GCS and BSO were also studied by Griffith. 24

Because

inhibition of γ-GCS by L-BSO occurred too rapidly (maximal inhibition within 1 minute) for kinetic measurements by conventional methods, DL-BSO was used for kinetic studies. Enzyme inhibition followed first order kinetics at any fixed concentration within the range (10-100 µM) of BSO examined.

The rate-limiting constant was 3.7 min,20 which corresponds to a T1/2 of

approximately 11s at a saturating concentration of inhibitor.

The maximal estimate of the

apparent initial binding constant for BSO was 100 µM. However, since BSO has two centers of asymmetry, one at the a-carbon and another at the sulfur, four isomers may possibly exist.21 In this study, it was assumed that only one isomer was active, thus the actual initial binding constant was extrapolated to be 25 µM. Extensive binding studies using column chromatography showed that extent of binding γGCS to BSO in the presence of ATP was examined using buffers with or without Mg and/or ATP to determine the role of these agents in the interaction of BSO and ?-GCS. Results of these studies indicated that 1 M equivalent of the BSO-phosphate adduct was bound tightly, noncovalently, and irreversibly to the fully inhibited enzyme.

?-GCS (725 U., 0.84 nmol was

inhibited to > 98% at 15 minutes in the presence of 0.83 mM L-BSO. However, when the enzyme was pre-incubated with cystamine, ?-GCS was not inactivated by subsequent exposure to BSO.

A possible explanation is that cystamine may bind to an enzyme sulfhydryl near the

active site of G-GCS. If the buffer contained Mg without ATP, γ-GCS-BSO binding was slightly

13

diminished, but the enzyme retained substantial (approximately 30%) activity.

When the buffer

contained neither Mg nor ATP, binding did not occur, and the enzyme retained full activity. In additional experiments using [?- 32 P]ATP or [8- 14 C]ATP, both radiolabels were bound to the enzyme inhibited with BSO.5

When the BSO-bound enzyme was analyzed by chromatography,

35

radiolabeled L-[ S]BSO or [?- 32 P]ATP coeluted distinctly from either BSO, P?, or ATP. The elution profile of the BSO-phosphate adduct indicated a net charge which was more negative than that of P?.

These observations were consistent with the profile expected for the BSO-

phosphate adduct.24 Other studies have attempted to elucidate the effects of BSO-induced GSH depletion on the activity of certain cytotoxic agents.

The mechanism through which GSH modulates

cytotoxicity is thought to be related to its role in cellular metabolism. Once γ-GCS is inactivated by BSO, GSH synthesis is inhibited, resulting in depletion of intracellular GSH.

Decreased

levels of GSH have been shown to increase the toxicity of various antineoplastic agents, in particular melphalan (L-PAM).6; 7; 25-27 Through

its

nucleophilic

and

oxidative-reductive properties, GSH may detoxify

potentially harmful reactive intermediates formed form oxidant injury, either by normal aerobic metabolism or through the action of xenobiotics.

GSH can react with electrophiles to form

thioethers, or, through its oxidation to GSSG, GSH can donate hydrogen to radical compounds to form the original molecule.

During normal steady-state conditions, GSH (0.5-10 mM) exists

primarily in its reduced form, but it may also exist in any of the following three interchanging states: 1) GSH may be oxidized by GSH peroxidase or by nonenzymatic means to a disulfide, GSGG (normally 5-50 µM), which may in turn be reduced to GSH by GSH reductase or by an exchange reaction;

2) GSH may form mixed disulfides with both protein and nonprotein

sulfhydryls; or, 3) GSH may exist in the form a thiol ester.28;

29

Peristeris et al.,30 investigated the

role of the antioxidative function of GSH in regulating tumor necrosis factor (TNF) production in vivo in lipopolysaccharide (LPS) treated mice.

TNF is a mediator in the pathogenesis of

endotoxic and septic shock and TNF inhibition has a protective effect in several animal models.

14

Pretreatment with GSH produced a significant inhibition of TNF production, whereas treatment with BSO had the reverse effect. Arrick and Nathan28 have classified the action of GSH on the therapeutic efficacy of antineoplastic agents as follows: 1) GSH may detoxify active species produced by agents such as L-PAM, CTX, nitrosoureas, radiation, and radiosensitizers or repair direct injury induced by such species through the role of GSH in oxidation-reduction and nucleophilic reactions; 2) GSH may increase the toxicity of certain agents such as bleomycin by participating in the formation of active cytotoxic species;

3) GSH may enhance the delivery of certain cytotoxics such as

methrotrexate (MTX) to their target site through its role in amino acid transport; or,

4)

alterations of GSH levels may interfere with the effectiveness or toxicity of concomitant or subsequent therapy.

Since the mechanism of the interaction between GSH and antineoplastics

has not been fully elucidated, the classification of certain drugs into these categories may be arguable. BSO has been effective both in vitro and in vivo in enhancing the cytotoxicity of L-PAM. However, the data from some studies have been conflicting as to the mechanism of the modulation of L-PAM cytotoxicity by GSH. Susukake et al, 31 studied the role of GSH in the modulation of L-PAM toxicity in L1210 L-PAM-sensitive and L1210R L-PAM resistant murine leukemia cell lines.

No differences in L-PAM uptake were noted between the resistant or

sensitive cell lines exposed to L-PAM 2.5 µg/ml for up to 45 minutes.

In addition, no

differences in L-PAM efflux were observed over 90 minutes after a 45-minute incubation of each cell line with L-PAM. However, marked differences between the two cell lines were noted with respect to the metabolism of L-PAM. In the L1210R cells exposed to 1.25 µg/ml of L-PAM for 3 hours, the ratio of unchanged L-PAM to its dechlorinated noncytotoxic derivative, dihydroxy LPAM

(4-[bis(2-hydroxyethyl)amino]-L-phenylalanine),

was

approximately

20:18.

The

corresponding ratio (approximately 24:7) in the L1210 cells was markedly different and indicated that dechlorination occurred to a lesser extent in the L-PAM-sensitive cells. When the L1210R cells were treated with the 0.1-0.25 mM N-ethylmaleimide (a sulfhydryl agent) 5 minutes prior to exposure to L-PAM, dechlorination of L-PAM was inhibited and occurred at the

15

same level as in the L1210 cells. Levels of the dihydroxy metabolite in either cell line correlated positively with GSH levels, which in turn correlated inversely with sensitivity to L-PAM. These observations led the authors to conclude that GSH-mediated resistance to L-PAM was probably not associated with a classical transport system in either L1210 cell line investigated, since neither accumulation nor retention of L-PAM correlated with drug resistance. Because of another report32 which indicated changes in transport of L-PAM as the mechanism of drug resistance, the authors conceded that such mechanisms may vary among cell lines examined.

Although the precise mechanism of GSH-mediated resistance to L-PAM remains

undefined, the authors have postulated that the interaction between L-PAM and GSH may be mediated by GSH-S-transferase, an enzyme which is involved in the dehalogenation of many electrophiles.31 Results of the study of Green et al6 with human ovarian cancer cells confirmed the findings of Suzukake, et al.31 No detectable difference was observed in the uptake of L-PAM (3.5 µM) between cells (1847me) with 4-fold resistance to L-PAM acquired in vivo and the parent cell line (1847) which was sensitive to L-PAM. There was a marked difference, however, in the cellular content of dihydroxy L-PAM which was 2-fold higher in the 1847me than in the 1847 cells. These observations led the authors to conclude that GSH-conferred resistance to L-PAM in this cell line may be related to increased catabolism of L-PAM to the noncytotoxic metabolite. Hamilton et al7 were also unable to detect any difference in L-PAM uptake in another L-PAMresistant human ovarian cancer cell line (2780ME) following exposure to 25 µM BSO. It was noted that the activity of GSH-transferase was higher in the 2780ME cells than in the L-PAMsensitive parent cells. Koberle and Speit 33 studied the effect of GSH depletion of sister chromatid exchange (SCE) induction by cytostatic drugs. V79 cells were treated with BSO for 24 hours either before BrdU application or together with BrdU.

The results indicate that BSO itself did not induce

SCEs either before BrdU treatment or together with BrdU . However, cellular depletion of GSH by BSO had a clear effect on SCE induction by doxorubicin and cyclophosphamide, with drug induced SCE frequencies were significantly increased after BSO pretreatment 16

Kang et al,34 compared the effects of cadmium on epidermal growth factor (EGF)induced DNA synthesis and its effects on GSH metabolism in quiescent NRK-49F cells with those of BSO.

Cadmium inhibited EFG-induced DNA synthesis in a dose-dependent manner

and increased cellular GSH content in growth arrested and EGF-stimulated NRK-49F cells. BSO also inhibited EGF-induced DNA synthesis but depleted cellular GSH content in EGF stimulated NRK-49F cells.

The BSO effects showed different dose dependencies.

The

combination of BSO and cadmium inhibited EGF induced DNA synthesis in NRK-49F cells in an additive manner, suggesting different mechanisms of action for the two agents. The results of this study demonstrate that cadmium inhibition of EGF-stimulated DNA synthesis is not mediated by GSH depletion. Possible mechanisms by which BSO inhibits cell proliferation were investigated using normal rat kidney fibroblasts (NRK-49F). NRK-49F cells were cultured in Dulbecco’s modified Eagle medium(DMEM) with and without glutamine. glutamine-deficient DMEM. the cells.

NRK-49F cells did not proliferate in the

Addition of glutamine into the medium stimulated proliferation of

Addition of BSO and glutamine inhibited the glutamine-stimulated cell proliferation in

a BSO dose-dependent fashion. Based on these results, glutamine is required for NRK-49F cell proliferation.

BSO-induced cytostasis is most likely due to inhibition of cellular uptake of

glutamine as well as other amino acids.34 It is important to recognize that inhibition of GSH synthesis with BSO likely provides an advantage in attacking multi-drug resistant tumor cells that extends beyond just the utilization of GSH which occurs with a variety of drugs.

Upregulation of GSH and GSH synthetic and

utlization enzymes is associated with drug resistance in a variety of tumor types.18;

35-40

An

increased ability to synthesize GSH potentially allows tumor cells to overcome depletion of GSH though increased utilization that would be overcome by blocking GSH synthesis with BSO. Indeed, that has been observed in neuroblastoma cell lines, pointing toward the importance of targeting γ-GSH with BSO.41

17

PRECLINICAL ACTIVITY GSH Depletion and Effect on L-PAM Cytotoxicity BSO will be evaluated in combination with L-PAM in the initial clinical trials. L-PAM is a bifunctional alkylating agent active in several malignancies, including ovarian cancer.

The

response rate of intravenous (IV) L-PAM 130 days) survivor. Mice treated with BSO alone had a MST of 43 days. For the 15 mice given BSO and

30

10 mg/kg of L-PAM, MST was approximately 120 days, and there were 3 long-term survivors. For mice treated in Experiment 2, the MST was 46 days for mice given water alone, 68 days for mice given water and 5 mg/kg of L-PAM, and > 125 days for those treated with BSO and 5 mg/kg of L-PAM. It is unclear why Ozols et al.45 observed an improved survival in tumor-bearing mice treated with L-PAM and BSO, whereas Kramer et al. 25 and Somfai-Relle et al.43 did not.

One

possible explanation is that the extent of GSH depletion achieved in the study by Ozols et al.45 was greater than that in the study of Kramer et al. 25 In the former, nude mice received drinking water with L-BSO 30 mM PO d x 5.

Five days following treatment with BSO, GSH was

depleted 79% in the bone marrow cells and 88% in the GI mucosa.45 In the study by Kramer et al.,25 BDF mice were given BSO 20 M PO d x 3 following a loading dose (450 mg/kg IP) or BSO 450 mg/kg IP every 6 h x 5. Six hours after the last dose of BSO, GSH was reduced by approximately 40% (BSO IP) in the bone marrow. Another possible explanation is use of the LBSO isomer specified by Ozols et al. in their study.45 whether the L isomer was used.25;

43

The other investigators did not specify

Ozols et al. also believed their in vivo system closely

represented an in vivo model of human ovarian cancer,45 and indeed, that ovarian xenograft was multi-drug resistant.

It is possible that multi-drug resistant tumors may show an advantage to

combining BSO with L-PAM that would not be seen when testing BSO and L-PAM in other systems, such as the L1210 leukemia system previously investigated.25; 43

BSO-Mediated Sens itization of Agents Other Than L-PAM Because GSH is potentially involved in modulating the activity of many antineoplastic agents, pre-clinical studies of BSO in combination with a variety of antineoplastic drugs and radiation have been conducted. Cyclophosphamide. Increased GSH as been associated with resistance to 4hydroperoxycyclophosphamide medulloblastoma

58

and leukemia

(4-HC, 59; 60

the

active

metabolite

of

cyclophosphamide)

in

cell lines, and in both of these systems, BSO enhanced 431

HC cytotoxicity. Similar results for BSO combined with other cyclophosphamide metabolites in leukemia cells have also been reported.61 It has been postulated GSH can diminish metabolism of 4-HC into metabolites that are more active at DNA alkylation. 62

BSO showed increased the

cytotoxicity of cyclophosphamide against pulmonary metastases in mice of the NFSa tumor, without increasing cytotoxicity to bone marrow in the mice.63 In the KHT saracoma,64 the MBT2 murine bladder cancer cells,65 and EMT6/SF tumors in mice,

66; 67

BSO was also able to

enhance the cytotoxicity of cyclophosphamide against the tumor cells. Dorr et al. studied the effect of BSO-mediated GSH depletion on the in vivo cytoxicity of the sulfhydryl-dependent drugs, BCNU, CTX, DOX, and L-PAM in BALB/c mice bearing MOPC-315 tumor cells.68

BSO at 50 mg/kg PO was administered 6 days after tumor

implantation, and the cytotoxics were administered at various doses one day following BSO. Twenty-four hours after BSO, GSH levels in the kidney or liver were reduced to < 10% of control values.

Pretreatment with BSO increased the % survival in mice by 20% at a BCNU

dose of 20 mg/kg and 25% at 35 mg/kg.

At low CTX doses (50 mg/kg), % survival as increased

by 37%; however, at higher doses (100-200 mg/kg) of CTX, % survival was decreased by 1621%. BSO produced a 15% increase in % survival at a DOX dose of 10 mg/kg, but there was no increase at 15 mg/kg of DOX. These results indicate that BSO may increase antitumor activity at lower doses of these agents, but alternatively, may increase adverse toxicity at higher doses. Subsequent work studying the effect of BSO on busulphan cytotoxicity against hematopoietic precursors in vitro did not show enhanced toxicity, again suggesting a tissue- or tumor-specific effect of BSO.69 BSO has been studied in combination with 4-HC in a rat brain tumor model in which both BSO and 4-HC were given by local installation in a polymer.70 Interestingly, local BSO delivery enhanced animal survival, whereas systemic BSO therapy did not. However, BSO was also

reported

to

diminish

the

protective

effect

of

challenging

cyclophosphamide (CTX) prior to higher-dose CTX treatment.71

mice

with

low-dose

The latter points toward the

possibility that sequence and timing of drug usage with BSO may effect the tolerability of the agent.

32

The mixed results with combining cyclophosphamide with BSO are likely due to the combination being active against tumors, but at the same time capable of systemic toxicity For example, BSO has been reported to increase urotoxicity of cyclophosphamide in mice.72;

73

.

Rapid cardiac toxicity (3 hours) has been reported when BSO was combined with cyclophosphamide in mice and rats.74 In vitro studies showed this to be a direct cytotoxic effect against cardiac myocytes, and also showed the ability to diminish the toxicity of combining 4HC + BSO by addition of thiol anti-oxidants, such as N-acetylcysteine or MESNA. 75 Platinum compounds. Increased detoxification through GSH pathways is thought to be a major mechanism of resistance to cisplatin (CDDP) and its analog carboplatin (CBDCA), the first line agents in ovarian cancer.

It has been shown that BSO can increase

sensitivity to platinum drugs in tumor cells cultured directly from individual patients,76 and in a similar study BSO enhanced cytotoxicity of cisplatin or doxorubicin.77 Lai et al. isolated lung cancer cells with acquired CDDP resistance by intermittent drug exposure.78 Enhanced DNA repair activity was seen in three cisplatin resistant cell lines. Pretreatment of the cells with BSO, followed by pentoxifyline (another resistance modulator) and CDDP combined treatment, effectively overcame acquired cisplatin resistance in vivo. The cytotoxicity of a 2 hr CDDP or tetraplatin (TP) exposure with or without a 24 hr BSO pretreatment was evaluated against human ovarian A2780 and murine leukemia L1210/0 cells and their corresponding CDDP resistant (A2780/CP and L1210/DDP).79 BSO significantly decreased the IC 50 values of CDDP and TP against both ovarian lines. BSO enhanced only the cytotoxicity of TP, but not CDDP against the L1210/0 cells, whereas it enhanced only the effect of CDDP against the L1210/DDP cells. BSO also increased platinum and L-PAM toxicity (but not that of doxorubicin or camptothean) for the U-937 leukemia cell line.80

Interestingly, in

cisplatin-resistant K562 leukemia cells, BSO enhanced NK cell killing, suggesting a role for GSH in the latter process.81 Kramer et al. suggested that the tissue selectivity of BSO-mediated GSH depletion may potentially determine which cytotoxic agents would be compatible for use with BSO.25 The

33

authors point to reports by Litterst et al. 82 and Kramer et al.,83 which indicated increased renal toxicity of CDDP at time of maximal GSH depletion by diethylmaleate, a thiol depleting agent, when used in combination with BSO.

Since BSO has been shown to have little effect on the

myelosuppression of L-PAM, Kramer et al. suggested that BSO might be more successfully combined with agents whose dose-limiting toxicity is myelosuppression.25 Radiation. Most of the data concerning the effect of BSO and radiation response have been generated from in vitro studies.

Biaglow et al. (1986) demonstrated that prolonged

exposure (24-120 hours) to 0.1 mM BSO increased, in a time-dependent fashion, the radiation response of A549 cells under normoxic conditions.84

An increase in single-strand breaks and

irreparable cross-linking of DNA was also noted. Following treatment with BSO and radiation, the increase in radiosensitivity was greater under normoxic than hypoxic conditions. The DMF was 2.4 under normoxic conditions. In contrast, Shrieve and Harris reported a small increase in radiation response in EMT6/SF cells treated with 50 µM BSO under hypoxic conditions and no increase in cells treated similarly under normoxic conditions.52 BSO has also been shown to increase the radiosensitizing effect of nitroimidazoles. Ono et al. examined the effect of BSO on radiation response to MISO in C3H/He mice bearing the NFSa tumor.85

Radiation response was assessed using tumor growth delay time.

BSO 5

mmole/kg IP produced a marked increase in the radiosensitizing effect of MISO 0.5 mmole/kg. The ER for MISO plus radiation, BSO plus radiation, and for BSO and MISO in combination with radiation was 1.44, 1.16, and 1.93, respectively. BSO has been useful in determining the role of GSH in intrinsic cellular resistance to heat and in the development of thermotolerance.

Shrieve et al. examined the effect of GSH-

depletion by BSO in heat-sensitive and heat-resistant HA-1 Chinese hamster fibroblast cell lines.86

Intracellular levels of GSH did not correlate with sensitivity to heat, but depletion of

GSH by BSO increased the sensitivity of both heat-sensitive and resistant cells to thermal stress. Furthermore, development of thermotolerance was inhibited in the heat-sensitive cells only when low levels of GSH were maintained for extended periods.

34

Other agents. As described above, BSO synergistically enhanced the cytotoxicity of the bioreductive agent tirapazamine in neuroblastoma cell lines.41 A number of other drugs, of diverse mechanisms, have been shown to have enhanced cytotoxicity for various tumor cell types when combined with BSO.

The benzocronycine deriviatve S29306-1, apparently an aklyating

agent, showed enhanced cytotoxicity when combined with BSO.87 Depletion of GSH by BSO in a glucocorticoid-resistant B-lineage ALL cell line restored the sensitivity to dexamethasone, apparently via a caspase-independent mechanism.88

The cytotoxic retinoid fenretinide, has been

shown to generate enhanced reactive oxygen species and apoptosis in leukemia cell lines when combined with BSO.89 BSO was able to overcome resistance to arsenic trioxide in leukemia cell lines,90 but the combination of arsenic trioxide and BSO was rapidly fatal to mice (Reynolds CP, et al, unpublished).

Dethylstilbesterol cytotoxicity for a protstate cancer cell line was enhanced

by BSO.91

35

PRECLINCIAL TOXICITY OF BUTHIONINE SULFOXIMINE (BSO, NSC-326231) ALONE AND IN COMBINATION WITH MELPHALAN (NSC-8806) (note, the following section is NCI data as reproduced from the original BSO Investigator’s Brochure)

SUMMARY: Buthionine sulfoximine (BSO NSC-326231) inhibits tissue glutathione (GSH) synthesis, lowers intracellular GSH levels and increases the efficacy of alkylating agents used in cancer chemotherapy.

Single intravenous and oral doses of BSO produced a reversible depletion of

GSH levels in plasma, lung, liver and kidney tissue in mice.

BSO treatment resulted in no

clinical signs of toxicity nor microscopic lesions in mice when given as single and multiple oral and intravenous doses ranging from 100 to 800 mg/kg.

Multiple intravenous doses of 1500

mg/kg dose of BSO (q4h x 6) produced a transient decrease in peripheral WBC in female mice. BSO produced no changes in other clinical pathology parameters and no gross or microscopic lesions were found in mice treated with 1600 mg/kg/dose (q4h x 6). A drug combination study was designed to address the effect of BSO on melphalan-induced toxicity. Multiple intravenous doses of 1600 mg/kg/dose (q4h x 6) of BSO resulted in maximal GSH depletion and caused a potentiation of melphalan-induced bone marrow, lymphoid and rental toxicity. Multiple oral and intravenous doses of up to 800 mg/kg/dose of BSO resulted in maximal liver GSH depletion and no potentiation of melphalan toxicity. BSO was lethal to dogs following a single intravenous dose of 3200 mg/kg/or up to 10 oral doses of 800 mg/kg/dose (q8h).

Clinical signs included emesis, diarrhea, hyperactivity,

tonic/clonic convulsions and seizures in the 800 mg/kg/dose group given up to 10 doses and emesis and diarrhea in the 400 and 100 mg/kg/dose groups given a total of 15 doses.

BSO

administration also resulted in a significant decrease of GSH concentration in whole blood, lung, liver and kidney.

Significant increases in SGPT, SGOT and alkaline phosphatase occurred in

dogs receiving a single intravenous dose of 3200 mg/kg or 9-10 oral doses of 800 mg/kg dose of BSO (q8h).

Dogs sacrificed in a moribund condition after a single intravenous dose of 3200

mg/kg of BSO had microscopic lesions in the urinary bladder and exocrine pancreas. 36

Fifteen

oral doses (q8h) of 100 mg/kg dose of BSO were well tolerated by beagle dogs. Multiple IV doses (60, 120 and 240 mg/kg/dose q 12 hours x 6) were also well tolerated in beagle dogs, although slight decreases in RBC, HGB, and HCT were noted.

EXPERIMENTAL PROCEDURES: Drug Preparation L-Buthionine-S, R-sulfoximine (BSO, NSC-326231) was formulated in sterile 0.9% sodium chloride injection.

Melphalan (L-Pam, NSC-8806) was supplied as the clinical

formulation. Sterile sodium chloride (0.9%) for injection was used as the vehicle (VCTL) in all studies. Animals Purebred beagle dogs used in these studies were supplied by the Developmental Therapeutics Program (DTP), Division of Cancer Treatment (DCT), and National Cancer Institute (NCI).

CD2F1 mice were supplied by the NCI through Charles River Breeding

Laboratories (Portage, MI). Bioavailability Study in Mice Male CD2F1 mice were administered either 0.9% sodium chloride for injection (VCTL) or 800 mg/kg of BSO orally or intravenously. Liver and plasma samples were collected at 0, 1, 2, 4, 8, 12, and 24 hours after dosing and assayed for GSH concentration. Blood samples for plasma BSO level analysis were collected at 0, 2, 5, 10, 20, 30, 60, 90 minutes and 2, 4, 8, 12, 16, 20 and 24 hours after dosing. In a second group of mice, lung, liver and kidney samples were collected 4 hours after BSO of VCTL administration and analyzed for GSH concentration.

37

Dose Response Study in Mice Male CD2F1 mice were assigned to 3 dose groups and a VCTL group (12 mice per group). Mice received a single dose of 1600, 1200, 800, or 0 mg/kg of BSO and were sacrificed 8 hours or 24 hours after dosing.

Liver and plasma samples were collected and analyzed for

GSH concentration. Multiple Dose Toxicity in Mice Oral Dose Study – Mice and female CD2F1 mice were assigned to 3 dose groups with a vehicle control (VCTL) group (21 male and 15 female per group). Mice received BSO or 0.9% sodium chloride (VCTL) orally by gavage every 8 hours for a total of 15 doses. BSO doses administered were 800,400,100 or 0 mg/kg/dose. Three mice of each sex in each dose group were sacrificed after 5 and 10 doses of BSO and on study days 9, 13, and 30. The mice were evaluated for adverse clinical signs, changes in clinical pathology parameters, gross and microscopic lesions. Additionally, 6 male mice per dose group were sacrificed 4 hours after the 15th dose of BSO and plasma, lung, liver and kidney GSH concentrations were determined. Intravenous Dose Study – Male and female CD2F1 mice were assigned to 3 dose groups and a VCTL group (30 male and 25 female mice per group). Mice were treated intravenously with BSO or 0.9% sodium chloride every 4 hours for a total of 6 doses. Doses were 1600, 800, 400 and 0 mg/kg/dose BSO. Tissues from five mice of each sex in each dose group were evaluated histologically on study day 2 and 29. Clinical pathology samples were taken on study day 2, 8, 15, 22, and 29. Additionally, 5 male mice per dose group were sacrificed one hour after the 5th dose of BSO and liver GSH concentration was determined.

38

Effect of BSO on Melphalan-Induced Toxicity in Mice Oral Dose Study – Male and female CD2F1 mice were assigned to 5 dose groups (25 male and 25 female mice per dose group) as follows:

Dose

Pretreatment

Treatment

Group

BSO (mg/kg/dose)

Melphalan (mg/kg)

A

800

5

B

800

1

C

800

0

D

0

5

E

0

1

BSO or pretreatment VCTL (0.9% sodium chloride) was administered orally by gavage every 8 hours for a total of 6 doses. One hour after the 5th dose of BSO or pretreatment VCTL, mice were treated with a single intravenous dose of melphalan or treatment VCTL. parameters were determined on study days 4, 5, 11, 18, and 30.

Clinical pathology

Gross and histopathology

examinations were done on days 11 and 30. Additionally, 5 mice of each sex were treated with 800 mg/kg/dose BSO or pretreatment VCTL every 8 hours and sacrificed for plasma, lung, liver and kidney GSH determinations one hour after the 5th dose of BSO. Intravenous Dose Study – Male and female CD2F1 mice were assigned to 3 BSO dose groups and once VCTL group (35 male and 30 female mice per group).

Mice were pretreated with

1600, 800, 400 mg/kg/dose of BSO or 0.9% sodium chloride (VCTL) intravenously every 4 hours for a total of 6 doses. One hour after the 5th dose of BSO, a single intravenous 5 mg/kg/dose of Melphalan was administered.

Clinical pathology parameters were determined on

39

days 2, 4, 6, 11, 17 and 30. Histopathological examinations was done on tissues collected on days 2 and 30. Liver glutathione levels were determined in 5 male mice from each dose group sacrificed one hour after the 5th dose of BSO. Bioavailability Study in Dogs Male beagle dogs were dosed with either 120 mg/kg or 0 mg/kg BSO intravenously (3 dogs per group). Blood samples were collected for plasma BSO and GSH analyses at 0, 25, 5, 10, 20, 30, 60, 90, 120 minutes and 4, 8, 10, 12, 16, 20 and 24 hours after dosing. Due to lack of GSH depletion, the dogs in the BSO group were treated with 3200 mg/kg BSO intravenously on study day 29 and blood was collected for plasma GSH analysis of 0, 1, 2, 4, 8, 12 and 24 hours after BSO. Complete necropsy and gross and microscopic examinations were done on all dogs sacrificed moribund on day 30.

In addition, liver, kidney and lung samples were collected for

tissue GSH analysis. Multiple Dose Toxicity Study in Dogs Male and female beagle dogs were assigned to 3 dose groups and VCTL group (4 dogs per group).

BSO doses (800, 400, 100n and 0 mg/kg/dose) were administered orally

approximately every 8 hours for a total of 15 doses. Blood was collected for whole blood GSH levels on study days 2, 4 and 8 and for clinical pathology determinations on days 2, 4, 8 and weekly thereafter to study day 65.

Liver GSH was assayed in all dogs sacrificed in a stress

moribund condition and all dogs sacrificed on study day 8.

Complete gross and microscopic

examinations were done on dogs sacrificed moribund and at scheduled necropsies on study day 8 and 65. RESULTS Pharmacokinetics BSO was rapidly cleared from the plasma of dogs and mice after single intravenous or oral doses of 2400 mg/m2 (Table 1). The plasma BSO concentrations versus time data is best

40

described as a bi-exponential function in mice and in 2 out of 3 dogs. In mice, the initial phase had t1/2 of approximately 5 minutes and accounted for 94% of the total AUC. The T1/2 of the terminal phase was approximately 40 minutes and accounted for 6% of the total AUC in mice. By contrast, dogs dosed with 2400 mg/m2 had an initial phase t1/2 of about 8 minutes and the initial phase represented only 20% of the total AUC. The t1/2 of the terminal phase was 36.3 minutes and accounted for 80% of the total AUC in dogs.

Even with the difference in BSO

elimination, mice and dogs had a similar rate of plasma drug clearance. Clearance of BSO from plasma was 84 ml/min/m2 and 104 ml/min/m2 in mice and dogs, respectively. Bioavailability Bioavailability of BSO given orally, based on plasma drug levels, appeared to be about 2% in mice, suggesting that BSO was either not absorbed from the gastrointestinal tract or that it was rapidly removed from the plasma. There was no apparent correlation between plasma BSO levels and tissue GSH depletion after oral or intravenous dosing.

Additionally, there was no

significant difference in the time course or magnitude of plasma and liver GSH depletion after oral and intravenous dosing with BSO. A single oral or intravenous dose of 800 mg/kg (2400 mg/m2 ) of BSO resulted in 75% depletion of plasma GSH and a 70% depletion of liver GSH. Maximal GSH depletion occurred in liver and plasma within 2-4 hours after intravenous and 4-8 hours after oral dosing. There is no significant difference between the GSH depletion after oral and intravenous dosing, which suggests that BSO is well absorbed from the gastrointestinal trace and is rapidly removed from the plasma into tissue compartments. Studies in Mice In the dose response study, mice were treated with a single oral dose of 1600, 1200 and 800 mg/kg of BSO. Eight hours after dosing, plasma GSH levels were depleted about 70% and liver GSH levels were depleted about 73%.

At 24 hours after dosing, plasma GSH

concentrations were 37 and 53% depleted, even though liver GSH had returned to 85-100% of control levels. This data suggests that maximum GSH depletion occurs in mice after a single 800 mg/kg oral dose of BSO.

41

Mice given BSO orally at doses of 800, 400, and 100 mg/kg every 8 hours for a total of 15 doses showed a significant decrease in plasma GSH at all dose levels although the depletion did not appear dose-dependent. Four hours after the 15th dose of BSO, plasma GSH levels were depleted by 61-72% compared to vehicle control values. A dose response was present in some tissues; four hours after the 15th dose of BSO, kidney GSH was depleted by 89%, 82% and 41% and lung GSH was depleted by 60%, 48% and 35% in the 800, 400, and 100 mg/kg/dose groups, respectively. Approximately 35% GSH depletion occurred in the livers of mice in the 800 and 400 mg/kg/dose groups, but there was no apparent dose response presumably due to rapid recovery of liver GSH levels after BSO administration. In spite of marked GSH depletion, mice treated orally with 800, 400, and 100 mg/kg dose for a total of 15 doses of BSO showed no adverse clinical signs, no changes in clinical pathology parameters, or gross or microscopic lesions at any dose level (Table 2). The only adverse sign of toxicity was a slight loss in body weight in male mice during the dosing period. Between study day 2 and 6, male mice lost an average of 10.7% of their day – 1 body weight in the 800 mg/kg/dose group. By study day 10, body weights of these mice were comparable with the VCTL group. BSO administered intravenously at doses of 1600, 800, and 400 mg/kg/dose (q 4h x 5) caused an 88% depletion of liver GSH in male mice. There was no apparent dose response in liver GSH depletion at the administered BSO dose levels; all 3 BSO doses produced maximal GSH depletion.

Toxicity of intravenously administered BSO was limited to a transient

depression (67%) in WBC levels and a slight 10% body weight loss in female mice in the 1600 mg/kg/dose group (Table 2).

There were no drug-related gross or microscopic lesions in any

mice treated with BSO intravenously (q 4h x 6) at doses up to 1600 mg/kg/dose. BSO given to mice on a chronic basis caused cataracts,92 but this particular toxicity has not seen in acute dosing of animals or humans. The effect of BSO on melphalan-induced toxicity was assessed in CD2F1 mice. A single intravenous dose of 5 mg/kg of Melphalan produced gastrointestinal toxicity and a 50% depression of peripheral WBC counts with histopathological evidence of thymic atrophy, but no evidence of bone marrow atrophy.

Pretreatment of mice with up to 800 mg/kg/dose of BSO 42

either intravenously (q 4h x 6) or orally (q 8h x 6) did not potentiate melphalan-induced leukopenia, however, BSO pretreatment appeared to slightly delay the recovery of WBC to control values.

In mice pretreated intravenously with 1600 mg/kg/dose of BSO (q 4h x 6),

melphalan produced a rise in blood urea nitrogen and the appearance of renal nephrosis in 5/5 male mice and 1/5 female mice.

In addition, BSO potentiated the bone marrow toxicity of

melphalan as evidenced by the appearance of histopathological evidence of bone marrow atrophy in 2/5 male mice.

Accompanying bone marrow atrophy, mice pretreated intravenously with

1500 mg/kg/dose of BSO before melphalan administration had splenic myeloid tissue atrophy and splenic lymphoid and mesenteric lymph node necrosis. All of these lesions were reversible and were completely resolved by day 30. Studies in Dogs A single intravenous dose of 3200 mg/kg of BSO was lethal to beagle dogs producing hyperactivity, repeated tonic/clonic convulsions, acute hemorrhage and ulceration of the urinary bladder with hematuria and mild to marked depletion of exocrine pancreas. Multiple intravenous doses (60, 120, and 240 mg/kg/dose) of BSO were administered every 12 hours x 6. No apparent hematologic changes occurred at the 60 mg/kg/ dose level. An incidental decrease in the neutorphil count of one dog occurred on days 5 and 36 (38% and 23% below baseline, respectively).

Drug-related hematologic changes which included a slight

depression of the RBC, HGG, and HCT were observed in one dog at the 120 mg/kg/dose level. Increases in neutrophil and lymphocyte were observed in three dogs at the same dose on several days throughout the study. One female exhibited an incidental decrease in platelet count (56% below baseline) in RBC, HGB, and HCT in two males on day 5. Two males and one female exhibited an incidental decrease in neutrophils on day 5 (28% below baseline in one male, 2440% in the other, and 300-36% in the female). All parameters had returned to baseline by the end of the study. There were no treatment-related changes in clinical chemistry parameters. In addition, there were no apparent drug-related gross necropsy observations or histiopathological lesions.

43

Multiple doses (800, 400 and 100 mg/kg/dose) of BSO were administered orally to dogs approximately every 8 hours for up to 15 doses.

Whole blood GSH levels were maximally

depleted by 52-70%. These changes in whole blood GSH, however, did not occur in a doserelated manner. Liver GSH levels in the 800 mg/kg/dose group were 95% depleted in the 3 dogs sacrificed moribund 6-24 hours after their 9th or 10th dose of BSO. Sixty hours after the 15th dose of BSO, dogs in the 400 mg/kg/dose group had a 78% depletion of liver GSH.

In the 100

mg/kg/dose group, liver GSH levels were comparable to control values 60 hours after the 15th dose of BSO. Dogs in the 800 mg/kg/dose group received only 9 or 10 doses group received only 9 or 10 doses of BSO and displayed adverse clinical signs of emesis, diarrhea, hyperactivity, and convulsions during the dosing period (Table 3).

Three out of four of these

dogs had multiple convulsions, so BSO dosing was discontinued and the dogs were sacrificed as moribund on study day 4 or 5 for humane reasons. scheduled sacrifice on study day 65.

The remaining dog survived until its

Increases in SGOT and alkaline phosphatase activities

occurred in all the dogs in the 800 mg/kg/dose group beginning on day 2. Emesis and diarrhea occurred in the 400 mg/kg/dose group beginning on study day 2 and continued during the dosing period. One out of four dogs in the 400 mg/kg/dose group was hyperactive after receiving the 8th dose of BSO, but no convulsive activity was present. Increased SGPT activity occurred in ¾ dogs in the 400 mg/kg/dose group and elevated SGOT and alkaline phosphatase occurred in one dog between study days 4 and 15. There were no drug-related microscopic lesions in any of the dose groups.

DISCUSSION: Buthionine sulfoximine (BSO, NSC-326231) is a potent inhibitor of t-glutamylcysteine synthetase, an enzyme involved n the synthesis of glutathione (GSH). tripeptide that acts to protect cells from a variety of cytotoxins.

GSH is an intracellular

Because of the chemically

reactive nature of most anti-neoplastic agents, GSH also protects tumor cells from the cytotoxicity of chemotherapeutic drugs used in the treatment of cancer.

Numerous pre-clinical

studies have shown that GSH depletion produced by BSO administration increases the efficacy 44

of melphalan, a potent alkylating agent, in human tumor cell lines and human tumor xenografts in nude mice. The present studies were designed to determine the toxicity of BSO in dogs and mice and additionally to explore the effect of BSO on melphalan-induced toxicity in mice. BSO

administration

caused

toxic

effects

gastrointestinal and urinary tracts of beagle dogs.

to

the

central

nervous

system

and

Clinical pathology changes indicated possible

hepatic toxicity (increases in SGOT, SGPT and alkaline phosphatase), however, no drug-related microscopic lesions were present in the liver. BSO administration caused severe GSH depletion in dogs, however, the condition of the dogs after multiple convulsions make quantitative interpretation of the changes in GSH levels difficult, so no direct comparison in the magnitude of GSH depletion and the toxicity of BSO in mice and dogs can be made. Toxicity was reversible and there was no apparent sex difference in the response of dogs to BSO. Mice tolerated large multiple oral and intravenous doses of up to 800 mg/kg/dose of BSO without apparent adverse effects. The only adverse effect of BSO in male mice receiving multiple oral doses of BSO was a significant but quantitatively small weight loss.

The administration of 1600 mg/kg/dose of

BSO (q4h x 6) intravenously caused a transient depression of WBC counts in female mice, however, no bone marrow lesions were present at necropsy. Pretreatment of mice with multiple oral or intravenous doses of up to 800 mg/kg/dose of BSO to induce GSH depletion did not significantly alter the toxicity of melphalan.

However, melphalan administration produced renal,

bone marrow and lymphoid toxicity in mice pretreated with very high intravenous doses of BSO (1600 mg/kg/dose q4h x 6).

This potentiation of melphalan toxicity by BSO pretreatment

doesn’t appear to be related to GSH depletion, since a quantitatively similar amount of GSH depletion occurred at lower doses of BSO without any potentiation of melphalan toxicity. The elimination of BSO from plasma fit a bi-exponential function in both dogs and mice. The elimination half-life values for the initial and terminal phases were similar in dogs and mice. However, in mice the AUC of the initial phase accounted for 94% of the total AUC while in dogs it accounted for only 20% of the total AUC. In dogs, elimination was represented by the terminal phase of the curve and in mice it was represented by the initial phase. This difference in the pattern of BSO elimination may, in part, account for the absence of toxic effects of BSO in

45

mice.

Oral bio-availability of BSO based on plasma drug levels was approximately 2%.

However, comparable GSH depletion occurred after intravenous and oral dosing, suggesting BSO is rapidly removed from plasma into tissue compartments and that the oral bio-availability of BSO is most likely higher than 2%.

46

TABLE 1 PHARMACOKINETIC PARAMETERS OF BUTHIONINE SULFOXIMINE SPECIES

BEAGLE DOGS

CD2F1 MICE

(mg/kg)

120

800

(mg/m2 )

2400

2400

(ml/min/kg)

5.18 ±0.53

28

(ml/min/m2)

104

84

t1/2

initial (min)

7.7

5

t1/2

terminal (min

36.3 ± 9.7

37

Vd term

(ml/kg)

266 ± 72

1490

Vd 55

(ml/kg)

233 ± 23

280

AUC Total

(µg/ml-min)

23800 ±2400

28500

Dose

C1

47

TABLE 2 CHARACTERISTICS OF BUTHIONINE SULFOXIMINE TOXICITY IN MICE

SCHEDULE Doses

ORAL (q8h x 15)

I.V. (q4h x 6)

800 mg/kg/dose

1500 mg/kg/dose

400 mg/kg/dose

800 mg/kg/dose

100 mg/kg/dose

400/mg/kg/dose

Mortality

•

none

•

none

Clinical Signs

•

none

•

none

Clinical Pathology

•

none

Body Weights

Histopathology

•

•

1600 mg/kg/dose •

day 2

•

1 WBC in females

1600 mg/kg/dose

none

no drug related lesions

48

•

day 2

•

10% decrease in females

•

no drug related lesions

TABLE 3

BUTHIONINE SULFOXIMINE TOXICITY IN DOGS

SCHEDULE

Dose(s)

SINGLE DOSE (IV)

MULTIPLE DOSES (Oral q8h) X 9/10 800 mg/kg/dose mg/kg/dose

3200 mg/kg

x 15 400

120 mg/kg Mortality Clinical Signs

100 mg/kg/dose 3 /4 800 mg/kg/dose x 9/10 dose 800 mg/kg/dose

3 /3 3200 mg/kg 3200 mg/kg • day 1 • hyperactivity/aggressive behavior •

• • • •

day 1 – 6 diarrhea/emesis hyperactivity convulsions

convulsions/seizures 400 and 100 mg/kg/dose • • •

Clinical Pathology

hematuria dilated pupils hypersalivation

3200 mg/kg

• •

day 2 – 15 diarrhea/emesis

800, 400 mg/kg/dose

• •

day 1 increases in

• •

day 4 – 15 increases in:

Alkaline phosphatase

Alkaline phosphatase

SGOT

SGOT

SGPT

SGPT

Serum Glucose Chloride Hematocrit Histopathology

3200 mg/kg •

no drug related lesions

urinary bladder: hemorrhage ulceration •

pancreas: depletion of

49

EXPERIENCE WITH BSO IN HUMANS BSO has been tested in phase I trials in adults, alone and in combination with L-PAM. The initial phase I studies tested 30 min infusions of BSO every 12 hours, alone (6 to 10 doses, escalation of BSO dosing from 1.5 to 13 g/m2 ), or in combination with 15 mg/m2 of L-PAM given as a bolus, alone or in combination with BSO.14;

16

alone to L-PAM given with BSO showed a

significant increase in leukopenia and

Comparsion of courses of L-PAM

thrombocytopenia when L-PAM was given with BSO compared to L-PAM alone.

To further

enhance the degree of GSH depletion at the time of L-PAM administration, another dose schedule was tested, with a loading dose of 3 g/m2 bolus over 30 minutes i.v. followed by a 24, 48, or 72 hour continuous infusion of BSO at 0.75 g/m2 /hr.15 The preferred dosing appeared to be the BSO 3 g/m2 bolus loading dose followed by 72 hours of 0.75 g/m2 /hr BSO with L-PAM (15 mg/m2 bolus at hour 48 of BSO infusion).15

The latter schedule of BSO alone produced

minimal systemic toxicity, but when combined with L-PAM caused substantial, but tolerable, myelosuppression. A pilot study in pediatrics tested the same schedule and also tested increasing the c.i.v. for BSO to 1.0 g/m2 /hr.10; 11

Human Pharamacokinetics The T ½ of BSO was < 2 hours when given as 30 min infusions every 12 hours,14; ~ 2.3 hours after continuous i.v. infusion.10; L-PAM AUC.10;

11; 15

11; 15

16

and

BSO may decrease clearance and increase the

Significant differences in clearance and t1/2 have been observed between

the R and S stereoisomers that comprise the clinical formulation of BSO.93 The pediatric BSO

50

and L-PAM (given with BSO) pharmacokientic data and a summary of that from the adult clinical trials is presented in Table 4.

TABLE 4 PHARMACOKINETCIS OF BSO AND L-PAM + BSO Dose level

Total BSO Css (µM)

1 2 1 vs. 2 Adult data

Dose level

(0.75 g/m2/hr x 72 hours) 94

1 2 1 vs. 2 Adult data 2

BSO CLss (ml/min/m2)

Total BSO t 1/2 hr

346 +/- 176 524 +/- 207 P = 0.02 465 +/- 189

Total BSO AUC (mM x hr) 26 +/- 12 41 +/- 16 P = 0.01 31 +/- 8

193 +/- 77 177 +/- 72 P = 0.08 136 +/- 45

2.3 +/- 1.2 2.2 +/- 1.1 P = 0.60 3.7

L-PAM Peak (µM) 3.02 +/- 0.3 3.32 +/- 1.2 P = 0.46 N/A

L-PAM AUC (mM x hr) 66 +/- 24 82 +/- 27 P = 0.15 53 +/- 24

L-PAM CLss (ml/min/m2) 252 +/- 88 204 +/- 83 P = 0.22 343 +/- 171

L-PAM t 1/2 hr 1.5 +/- 0.5 1.9 +/- 0.5 P = 0.13 1.1 +/- 0.4

(15 mg/m L-PAM + q 12 hr BSO)

Pharmacodynamic studies showed both decreased GSH (40% of baseline) and inhibition of the γ-GCS activity (the enzyme targeted by BSO) were demonstrated in blood mononuclear cells.14; 16

A high degree of GSH depletion in tumor tissue (< 10% of baseline) during

continuous infusion BSO was demonstrated in a subsequent study.15 Depletion of GSH by BSO in blood mononuclear cells occurred to 30-40% of baseline in adults and ~46% in children, and appears to return to normal within 4 days of stopping the BSO. Pharmacodynamic modeling of data from a phase I trial suggested a gradual depletion of GSH occurring over about 30 hours, 95

51

and RNA expression of γ-GCS in mononuclear cells, which showed a 3-fold variability from patient to patient, also showed a compensatory upregulation in response to BSO treatment.80

Safety and Efficacy

The reported systemic toxicities of BSO and L-PAM in adults were unremarkable with reversible myelosuppression being the most common.14-16

Documented responses have occurred

in patients with melanoma (complete response 18 months), ovarian cancer, breast cancer, and small cell carcinoma of the lung.14-16 A pilot study of BSO combined with 15 mg/m2 of L-PAM in children with recurrent/refractory high-risk neuroblastoma achieved a 29% response rate (duration 1-3 months), with 7 partial (6 of 7 had previous myeloablative therapy) and 2 minor responses (Preprint in Appendix).10; relapse therapy.

11

Of 9 responders, 5 were treated with BSO/L-PAM as their primary

One patient, whose pre-relapse therapy only included chemotherapy with no

myeloablative therapy or 13-cis-retinoic acid, had a marked response to BSO/L-PAM with 98% reduction in size of left pelvic mass, improvement of liver nodules and resolution of bone marrow disease following 2 courses of therapy. There were 7 patients that experienced a greater than 25% improvement of their disease and most patients had a maximal response after 1 course of BSO/L-PAM therapy.

Three patients had multi-log improvement in bone marrow

involvement by immunocytology that corresponded to a marked reduction of tumor observed by routine microscopic examination of biopsy and aspirate. An additional patient showed resolution of bone marrow (tumor undetectable by immunocytology) of a marrow showing 50% tumor involvement by light microscopy prior to BSO/L-PAM therapy 52

(no immunocytology available

at baseline). Two patients had painful, subcutaneous scalp masses that completely resolved, one of these patients required continuous morphine that was no longer required within days after the 1st course of BSO/L-PAM.

Except for two patients with severe toxicity (see below), toxicity in the pediatric study was similar to that in adults, with most patients showing grade 3-4 leukopenia and thrombocytopenia. The degree of hematopoietic toxicity seen with BSO + L-PAM appears to be increased over what would be expected for L-PAM alone at the same dose. There appeared to be cumulative hematopoietic toxicity for patients treated with more than 2 courses of BSO + LPAM, especially with respect to platelet recovery. In contrast to the adult studies, there were 2 deaths due to CNS toxicity in the first pediatric study of BSO + L-PAM.

Both patients having a toxic death presented with acute

tubular necrosis; one had a large, intracranial mass.

No other significant toxicities were seen.

The first patient experiencing a toxic death had received one course of BSO + L-PAM without undue toxicity, and developed renal toxicity followed by neurotoxicity during the BSO infusion in the 2nd course; L-PAM was not given during the course of therapy with toxicity. The second patient (patient with the CNS mass) received both BSO and L-PAM before developing renal and neurotoxcity.

The onset of toxicity in the first patient during BSO alone might suggest an

unusual sensitivity to GSH depletion, but the patient had received BSO + L-PAM one month before without any remarkable toxicity. Review of all data from the two toxic deaths suggests the possibility of a drug interaction between BSO and cephalosporin antibiotics could have lead to BSO toxicity (preprint in appendix).11

Clinical review of both cases found that cephalosporin antibiotics were

53

administered during the period of BSO infusion (treatment of cellulitis in first patient; fever in second).

At least one other patient is known to have received cephalosporin (treatment of

cellulitis) during their first of three courses of BSO and had no renal or neurologic sequelae. An analysis of the pediatric pilot trial found that 14/32 patients had a history of either total body irradiation (TBI) (N=7), abdominal radiation (kidneys in field; N=6 not including TBI patients), or cranial radiation (N=6; not including TBI patients) prior to BSO therapy. 11

Of the 2

toxic deaths, one (pt #6) had previous TBI, whereas the other (pt. #30) had radiation to a nonspecified pelvic site (no kidney or cranial radiation).

Assessment of renal function during the

first course of therapy revealed 19/32 patients with no abnormalities in urinalysis or serum creatinine while receiving BSO, 4 patients had pre-existing 1-3+ proteinuria/glucosuria with normal serum creatinine, and 4 patients developing 1-2+ proteinuria/glucosuria with normal serum creatinine while on BSO.

The new onset during BSO therapy of > 3+ proteinuria and

glucosuria, combined with elevation of baseline serum creatinine, had a significant (P = 0.02 by Fisher’s exact test) correlation with BSO toxicity leading to death. The complete etiology of the 2 toxic deaths remains to be defined, as adult trials have had no reports of significant neurologic or renal toxicity.

Pre-clinical studies of BSO in animals

demonstrated minimal GSH depletion in brain (> 70% control) with rapid recovery13 and no neurologic toxicity except that when substantial doses of BSO were given to large animals, reversible aggressiveness and convulsions were observed.12-14

It is speculated that the large

intracranial mass contributed to the second pediatric patient’s grade 5 toxicity, perhaps in addition to the mass effect, by altering the blood-brain barrier and increasing intracranial drug levels.

54

Although descriptive, the analysis of prior radiation therapy in the 32 children treated with BSO suggested that there is no evidence of an association between a history of radiation prior to BSO therapy and the subsequent development of neurologic or renal toxicity. 11

Despite

no difference in plasma creatinine, pre-clinical evaluation of BSO has shown sub-clinical evidence of acute tubular necrosis by light microscopy in mice treated with BSO.96 Although early cephalosporins such as cephaloridine are known to cause acute tubular necrosis by a GSH-dependent process involving free radicals made during proximal renal tubular transport,97 the third generation cephalosporins used by the children experiencing grade 5 toxicity have a very low potential for proximal renal tubular membrane transport.97;

98

However,

even forms of cephalosporins considered safe for patients with renal failure have had anecdotal reports of acute tubular necrosis mediated by hypersensitivity reactions.99 Importantly, in this group of children treated with BSO, the new onset of high-grade proteinuria and glucosuria appears to foreshadow life-threatening complications. Although high-dose BSO has been reported to cause substantial nigral changes in rodents,100 review of the literature revealed no definite cause of these renal and neurologic toxicities. Neurotoxicity from BSO in cell culture has also been associated in vitro with increased extracellular copper levels,101 though there was no known reason why increased copper would have occurred in the patients with BSO toxicity. Until the exact etiology of BSO toxicity is defined, the authors recommend patients with intracranial disease be excluded and all cephalosporin antibiotics be avoided during BSO administration. It is also important that drugs known to deplete GSH, such as acetaminophen, not be given just prior to, together with, or soon after BSO administration.

55

SUMMARY OF DATA AND GUIDANCE FOR THE INVESTIGATOR Data obtained with a variety of tumor systems have established a role for GSH in mediating drug resistance to several types of drugs, in particular alkylating agents and platinum compounds.

Increased glutathione (GSH) is common in drug-resistant cancer cells, and the

mechanism appears to be transcriptional up-regulation of RNA for, and consequently increased production of γ-glutamylcysteine synthetase

(γ-GCS), the initial enzyme in the synthetic

pathway for GSH. BSO is selective inhibitor of γ-GCS that can decrease GSH levels in tumor cells, prevent tumor-cell reactionary GSH increases in response to chemotherapy, and appears to have preferential activity in depleting GSH from tumor rather than many normal cells (including bone marrow).

BSO has been shown to be tolerable when given alone, and in combination with the alkylating agent melphalan (L-PAM) to animals and to humans, including children.

BSO

increases the degree of hematopoietic toxicity seen with L-PAM, limiting the use of L-PAM to low levels when combined with BSO. Because the dose-limiting systemic toxicity of BSO when combined with L-PAM is hematopoietic, increased activity may be achieved if extra-medullary toxicities allow dose escalation of L-PAM when given with BSO by using hematopoietic stem cell support.

Two toxic deaths (renal and neurotoxicity) occurred in the pediatric BSO + L-PAM pilot study, one associated with BSO treatment alone, the other with BSO + L-PAM. The etiology of those two toxic deaths remains unclear, though careful review of the two toxic deaths suggests

56

that an intracranial mass in one patient, and the possibility of an interaction between BSO and cephalosporin

antibiotics

pre-disposed,

or

perhaps

caused,

the

observed

toxicities.

Accordingly, investigators are cautioned that use of BSO in patients with intracranial mass disease, or in patients being treated with any potentially nephrotoxic or neurotoxic drug (including cephalosporin antibiotics) is contraindicated until further information on the potential for BSO toxicity in those situations is obtained.

Other drugs known to interact with GSH,

especially acetaminophen, are also contraindicated during the use of BSO.

The clinical activity seen with BSO + low-dose L-PAM in recurrent neuroblastoma suggests that BSO + L-PAM is a promising novel therapy for recurrent neuroblastoma, especially if L-PAM levels can be dose-escalated using hematopoietic stem cell support.

While

clinical studies have only been conducted with BSO combined with melphalan, it is possible that BSO may enhance the activity of other anti-neoplastic drugs.

However, there remains the

significant concerns for increased systemic toxicity, especially with agents such as the platinum compounds.

Thus, further pre-clinical toxicology studies will be necessary before undertaking

any clinical trials with combinations other than BSO + L-PAM.

In the case of apparent BSO toxicity, N-acetylcysteine (NAC) is not currently recommended, as the BSO will prevent NAC from enhancing GSH synthesis and it is unknown if the oral formulation of NAC available in the USA can be safely given in such a setting. An alternative would be the parenteral thiol agent sodium thiosulfate (STS), but again caution should be employed as the effects of combining STS with BSO are not well understood.

57

REFERENCES 1. Calcutt G, Connors TA . Tumoour sulphydryl levels and sensitivity to the nitrogen mustard merophan. Biochem Pharmacol. 1963;12:839-45.:839-845. 2. Tew KD, Houghton PJ, Houghton JA . Preclinical and Clinical Modulation of Anticancer Drugs. In: Anonymous. Preclinical and Clinical Modulation of Anticancer Drugs. Boca Raton: CRC Press; 1993:13-17. 3. Meister A , Anderson ME. Glutathione. Annu Rev Biochem. 1983;52:711-60:711-760. 4. Russo A , DeGraff W , Friedman N, Mitchell JB. Selective modulation of glutathione levels in human normal versus tumor cells and subsequent differential response to chemotherapy drugs. Cancer Res . 1986;46:2845-2848. 5. Simonian NA , Coyle JT. Oxidative stress in neurodegenerative diseases . Annu Rev Pharmacol Toxicol. 1996;36:83-106:83-106. 6. Green JA , Vistica DT, Young RC, Hamilton TC, Rogan AM, Ozols RF. Potentiation of melphalan cytotoxicity in human ovarian cancer cell lines by glutathione depletion. Cancer Res . 1984;44:54275431. 7. Hamilton TC, Winker MA , Louie KG, et al. Augmentation of adriamycin, melphalan, and cisplatin cytotoxicity in drug-resistant and -sensitive human ovarian carcinoma cell lines by buthionine sulfoximine mediated glutathione depletion. Biochem Pharmacol. 1985; 34:2583-2586. 8. Anderson CP, Tsai JM, Meek WE, et al. Depletion of glutathione by buthionine sulfoxine is cytotoxic for human neuroblastoma cell lines via apoptosis . Exp Cell Res . 1999;246:183-192. 9. Anderson CP, Tsai JM, Chan WW, et al. Buthionine sulfoximine (BSO) is cytotoxic and enhances the activity of melphalan for human neuroblastoma cell lines . Eur J Cancer. 1997;33:2016-2019. 10. Anderson CP, Seeger RC, Bailey HH, Reynolds CP. Pilot of buthionine sulfoximine (BSO) combined with non-myeloablative melphalan (L-PAM) against refractory neuroblastoma (NB) [abstract]. Proc Amer Soc Clin Oncol. 2002;21:298a . 11. Anderson CP, Perentis J, Neglia J, et al. Pilot of buthionine sulfoximine (BSO) combined with non-myeloablative melphalan (L-PAM) against refractory neuroblastoma (NB). Submitted. 2004. 12. Lee FY, Allalunis Turner MJ, Siemann DW . Depletion of tumour versus normal tissue glutathione by buthionine sulfoximine. Br J Cancer. 1987;56:33-38. 13. Terradez P, Asensi M , Lasso de la Vega MC, Puertes IR, Vina J, Estrela JM. Depletion of tumour glutathione in vivo by buthionine sulphoximine: modulation by the rate of cellular proliferation and inhibition of cancer growth. Biochem J. 1993;292:477-483. 14. Bailey HH, Mulcahy RT , Tutsch KD, et al. Phase I clinical trial of intravenous L-buthionine sulfoximine and melphalan: an attempt at modulation of glutathione. J Clin Oncol. 1994;12:194-205.

58

15. Bailey HH, Ripple G, Tutsch KD, et al. Phase I study of continuous-infusion L-S,R-buthionine sulfoximine with intravenous melphalan. J Natl Cancer Inst. 1997;89:1789-1796. 16. O'Dwyer PJ, Hamilton TC, LaCreta FP, et al. Phase I trial of buthionine sulfoximine in combination with melphalan in patients with cancer. J Clin Oncol. 1996;14:249-256. 17. Bailey HH. L-S,R-buthionine sulfoximine: historical development and clinical issues . ChemicoBiological Interactions. 1998;111-112:239-254. 18. Calvert P, Yao KS, Hamilton TC, O'Dwyer PJ. Clinical studies of reversal of drug resistance based on glutathione. [Review] [38 refs]. Chemico-Biological Interactions. 1998;111-112:213-224. 19. Hamilton D, Fotouhi-Ardakani N, Batist G. The glutathione system in alkylator resistance. [Review] [114 refs]. Cancer Treatment & Research. 2002;112:67-87. 20. Meister A . Inhibition of glutamine synthesis and gamma-glutamylcysteine synthetase by methionine sulfoximine and related compounds. In: Seiler P, Jung MJ, Koch WJ, eds. Enzyme-Activated Irreversible Inhibitors . Elsevier/North-Holland Biomedical Press; 1978:187-210. 21. Griffith OW . Depletion of glutathione by inhibition of biosynthesis . Methods Enzymol. 1981;77:59-63.:59-63. 22. Griffith OW , Anderson ME, Meister A . Inhibition of glutathione biosynthesis by prothionine sulfoximine (S-n-propyl homocysteine sulfoximine), a selective inhibitor of gamma-glutamylcysteine synthetase. J Biol Chem. 1979;254:1205-1210. 23. Griffith OW , Meister A . Potent and specific inhibition of glutathione synthesis by buthionine sulfoximine (S-n-butyl homocysteine sulfoximine). J Biol Chem. 1979;254:7558-7560. 24. Griffith OW . Mechanism of action, metabolism, and toxicity of buthionine sulfoximine and its higher homologs, potent inhibitors of glutathione synthesis . J Biol Chem. 1982;257:13704-13712. 25. Kramer RA , Greene K, Ahmad S, Vistica DT. Chemo sensitization of L-phenylalanine mustard by the thiol-modulating agent buthionine sulfoximine. Cancer Res . 1987;47:1593-1597. 26. Ozols RF, Hamilton TC, Young RC. Modulation of drug resistance in human ovarian cancer: experimental results and clinical applications. Prog Clin Biol Res . 1986;223:173-85.:173-185. 27. Suzukake K, Petro BJ , Vistica DT. Reduction in glutathione content of L-Pam resistant L1210 cells confers drug sensitivity. Biochem Pharmacol. 1982;31:121-124. 28. Arrick BA , Nathan CF. Glutathione metabolism as a determinant of therapeutic efficacy: a review. Cancer Res . 1984;4224-4232. 29. Meister A . Roles and functions of glutathione. Biochem Soc Trans. 1982;10:78-79. 30. Peristeris P, Clark BD, Gatti S, et al. N-acetylcysteine and glutathione as inhibitors of tumor necrosis factor production. Cell Immunol. 1992;140:390-399.

59

31. Suzukake K, Vistica BP, Vistica DT. Dechlorination of L-phenylalanine mustard by sensitive and resistant tumor cells and its relationship to intracellular glutathione content. Biochem Pharmacol. 1983;32:165167. 32. Redwood WR, Colvin M . Transport of melphalan by sensitive and resistant L1210 cells . Cancer Res . 1980;40:1144-1149. 33. Koberle B, Speit G. The effect of glutathione depletion on sister-chromatid exchange induction by cytostatic drugs. Mutat Res . 1990;243:225-231. 34. Kang YJ , Enger MD. Cadmium inhibits EGF-induced DNA synthesis but increases cellular glutathione levels in NRK-49F cells . Toxicology. 1991;66:325-333. 35. Gouaze V, Mirault ME, Carpentier S, Salvayre R, Levade T, Andrieu-Abadie N. Glutathione peroxidase-1 overexpression prevents ceramide production and partially inhibits apoptosis in doxorubicin-treated human breast carcinoma cells . Molecular Pharmacology. 2001;60:488-496. 36. Yang B, Keshelava N, Wan Z, et al. Glutathione (GSH) and RNA expression of GSH synthetic and utilization enzymes are increased in drug-resistant neuroblastoma cell lines [abstract]. Proc Amer Assoc Canc Res . 2003;44:1273. 37. Bai F, Nakanishi Y, Kawasaki M, et al. Immunohistochemical expression of glutathione Stransferase-Pi can predict chemotherapy response in patients with nonsmall cell lung carcinoma . Cancer. 1996;78:416-421. 38. Bailey HH, Gipp JJ, Mulcahy RT . Increased expression of gamma-glutamyl transpeptidase in transfected tumor cells and its relationship to drug sensitivity. Cancer Lett. 1994;87:163-170. 39. Mulcahy RT , Untawale S, Gipp JJ. Transcriptional up-regulation of gamma-glutamylcysteine synthetase gene expression in melphalan-resistant human prostate carcinoma cells . Mol Pharmacol. 1994;46:909914. 40. Cheng XS, Kigawa J, Minagawa Y, et al. Glutathione S-transferase-p expression and glutathione concentration in ovarian carcinoma before and after chemotherapy. Cancer. 1997;79:521-527. 41. Yang B, Keshelava N, Anderson CP, Reynolds CP. Antagonism of buthionine sulfoximine cytotoxicity for human neuroblastoma cell lines by hypoxia is reversed by the bioreductive agent tirapazamine. Cancer Res . 2003;63:1520-1526. 42. Nevin JE, III, Pinzon G, Baggerly TJ, Turner FG, Marsella JJ. The use of intravenous phenylalanine mustard followed by supervoltage irradiation in the treatment of carcinoma of the ovary . Cancer. 1983;51:1273-1283. 43. Somfai Relle S, Suzukake K, Vistica BP, Vistica DT. Reduction in cellular glutathione by buthionine sulfoximine and sensitization of murine tumor cells resistant to L-phenylalanine mustard . Biochem Pharmacol. 1984;33:485-490. 44. Skapek SX, Colvin OM , Griffith OW , et al. Buthionine sulfoximine-mediated depletion of glutathione in intracranial human glioma-derived xenografts . Biochem Pharmacol. 1988;37:4313-4317.

60

45. Ozols RF, Louie KG, Plowman J, et al. Enhanced melphalan cytotoxicity in human ovarian cancer in vitro and in tumor-bearing nude mice by buthionine sulfoximine depletion of glutathione. Biochem Pharmacol. 1987;36:147-153. 46. Kable EP , Favier D, Parsons PG. Sensitivity of human melanoma cells to L-dopa and DLbuthionine (S,R)-sulfoximine. Cancer Res . 1989;49:2327-2331. 47. Felix CC, Hyde JS, Sarna T, Sealy RC. Melanin photoreactions in aerated media: electron spin resonance evidence for production of superoxide and hydrogen peroxide. Biochem Biophys Res Commun. 1978;84:335-341. 48. Hansson J, Edgren M , Ehrsson H, Ringborg U, Nilsson B. Effect of D,L-buthionine-S,Rsulfoximine on cytotoxicity and DNA cross-linking induced by bifunctional DNA-reactive cytostatic drugs in human melanoma cells . Cancer Res . 1988;48:19-26. 49. Anderson CP, Keshelava N, Satake N, Meek WH, Reynolds CP. Synergism of buthionine sulfoximine and melphalan against neuroblastoma cell lines derived after disease progression. Med Pediatr Oncol. 2000;35:659-662. 50. Anderson CP, Seeger RC, Statake N., et al. Buthionine sulfoximine and myeloablative concentrations of melphalan overcome resistance in a melphalan-resistant neuroblastoma cell line. J Pediatr Hematol Oncol. 2001;23:500-505. 51. Anderson CP, Reynolds CP. Syngergistic cytotoxicity of buthionine sulfoximine (BSO) and intensive melphalan (L-PAM) for neuroblastoma cell lines established at relapse after myeloablative therapy. Bone Marrow Transplantation. 2002;30:135-140. 52. Shrieve DC, Harris JW . Effects of glutathione depletion by buthionine sulfoximine on the sensitivity of EMT6/SF cells to chemotherapy agents or X radiation. Int J Radiat Oncol Biol Phys. 1986;12:11711174. 53. Vukovic V, Nicklee T, Hedley DW . Differential effects of buthionine sulphoximine in hypoxic and non-hypoxic regions of human cervical carcinoma xenografts . Radiotherapy & Oncology. 2001;60:69-73. 54. Tsuji K, Gopalan P, Lehmann K, et al. Species and sex differences of aflatoxin B1-induced glutathione S-transferase placental form in single hepatocytes . Cancer Lett. 1992;66:249-254. 55. Ahmad S, Mulberg A , Aljian J, Vistica DT. Hepatic-mediated elevation and maintenance of metastatic tumor cell glutathione. Biochem Pharmacol. 1986;35:1697-1701. 56. Skapek SX, Colvin OM , Griffith OW , Elion GB, Bigner DD, Friedman HS. Enhanced melphalan cytotoxicity following buthionine sulfoximine-mediated glutathione depletion in a human medulloblastoma xenograft in athymic mice. Cancer Res . 1988;48:2764-2767. 57. Soble MJ, Dorr RT . Lack of enhanced antitumor efficacy for L-buthionine sulfoximine in combination with carmustine, cyclophosphamide, doxorubicin or melphalan in mice. Anticancer Res . 1988;8:1722.

61

58. Friedman HS, Colvin OM , Kaufmann SH, et al. Cyclophosphamide resistance in medulloblastoma . Cancer Res . 1992;52:5373-5378. 59. Peters RH, Jollow DJ, Stuart RK. Role of glutathione in the in vitro synergism between 4hydroperoxy -cyclophosphamide and cisplatin in leukemia cell lines . Cancer Res . 1991;51:2536-2541. 60. Peters RH, Ballard K, Oatis JE, Jollow DJ, Stuart RK. Cellular glutathione as a protective agent against 4-hydroperoxycyclophosphamide cytotoxicity in K-562 cells . Cancer Chemotherapy & Pharmacology. 1990;26:397-402. 61. Crook TR, Souhami RL, Whyman GD , McLean AE. Glutathione depletion as a determinant of sensitivity of human leukemia cells to cyclophosphamide. Cancer Res . 1986;46:5035-5038. 62. Lee FY. Glutathione diminishes the anti-tumour activity of 4-hydroperoxycyclophosphamide by stabilising its spontaneous breakdown to alkylating metabolites . Br J Cancer. 1991;63:45-50. 63. Ono K, Komuro C, Tsutsui K, et al. Combined effect of buthionine sulfoximine and cyclophosphamide upon murine tumours and bone marrow. Br J Cancer. 1986;54:749-754. 64. Richardson ME, Siemann DW . Thiol manipulation as a means of overcoming drug resistance in a novel cyclophosphamide-induced resistant cell line. International Journal of Radiation Oncology, Biology, Physics . 1992;22:781-784. 65. Lee KE, Mayer RD, Cockett AT. Effect of systemic glutathione depletion by buthionine sulfoximine on sensitivity of murine bladder cancer to cytotoxic agents . Urology. 1989;34:376-380. 66. Ono K, Shrieve DC. Effect of glutathione depletion by L-buthionine sulfoximine on the cytotoxicity of cyclophosphamide in single and fractionated doses to EMT6/SF mouse tumors and bone marrow. J Natl Cancer Inst. 1987;79:811-815. 67. Ono K, Shrieve DC. Enhancement of EMT6/SF tumor cell killing by mitomycin C and cyclophosphamide following in vivo administration of buthionine sulfoximine. International Journal of Radiation Oncology, Biology, Physics . 1986;12:1175-1178. 68. Dorr RT , Soble MJ, Greene D. Effect of glutathionine (GSH) depletion by L-buthionine (S/R) sulfoximine (BSO), on the toxicity and antitumor effacy of carmustine BCNU, cyclophosphamide (CTX), doxorubicin (DOX), and melphalan (L-PAM), in murine MOPC-315 plasmacytoma [abstract]. Proc AACR. 1986;27:374. 69. Hassan Z, Hellstrom-Lindberg E, Alsadi S, Edgren M , Hagglund H, Hassan M . The effect of modulation of glutathione cellular content on busulphan-induced cytotoxicity on hematopoietic cells in vitro and in vivo. Bone Marrow Transplantation. 2002;30:141-147. 70. Sipos EP , Witham TF, Ratan R, et al. L-buthionine sulfoximine potentiates the antitumor effect of 4-hydroperoxycyclophosphamide when administered locally in a rat glioma model. Neurosurgery . 2001;48:392400.

62

71. Carmichael J, Friedman N, Tochner Z, et al. Inhibition of the protective effect of cyclophosphamide by pre-treatment with buthionine sulfoximine. International Journal of Radiation Oncology, Biology, Physics . 1986;12:1191-1193. 72. Ishikawa M , Takayanagi Y, Sasaki K. Modification of cyclophosphamide-induced urotoxicity by buthionine sulfoximine and disulfiram in mice. Research Communications in Chemical Pathology & Pharmacology. 1989;65:265-268. 73. Ishikawa M , Sasaki K, Takayanagi Y. Injurious effect of buthionine sulfoximine, an inhibitor of glutathione biosynthesis, on the lethality and urotoxicity of cyclophosphamide in mice. Japanese Journal of Pharmacology. 1989;51:146-149. 74. Friedman HS, Colvin OM , Aisaka K, et al. Glutathione protects cardiac and skeletal muscle from cyclophosphamide-induced toxicity. Cancer Res . 1990;50:2455-2462. 75. Dorr RT , Lagel K. Effect of sulfhydryl compounds and glutathione depletion on rat heart myocyte toxicity induced by 4-hydroperoxycyclophosphamide and acrolein in vitro . Chemico-Biological Interactions. 1994;93:117-128. 76. Sargent J, Taylor C, Elgie A . In vitro modification of drug resistance in fresh tumor cells from patients with ovarian cancer. American Drugs. 1994;5:61. 77. Lewandowicz GM , Britt P, Elgie AW, et al. Cellular glutathione content, in vitro chemoresponse, and the effect of BSO modulation in samples derived from patients with advanced ovarian cancer. Gynecologic Oncology. 2002;85:298-304. 78. Lai SL, Hwang J, Perng RP, Whang-Peng J. Modulation of cisplatin resistance in acquiredresistant nonsmall cell lung cancer cells . Oncol Res . 1995;7:31-38. 79. Morita M , Prasad SB, Khokhara AR. Cytotoxicity of buthionine sulfoximine and its differential effects on the activities of cisplatin and tetraplatin against two pairs of sensitive-resistant cell lines [abstract]. Proc Annu Meet Am Assoc Cancer Res . 1995; 36. 80. Yao K, Godwin AK, Ozols RF, Hamilton TC, O'Dwyer PJ. Variable baseline gammaglutamylcysteine synthetase messenger RNA expression in peripheral mononuclear cells of cancer patients, and its induction by buthionine sulfoximine treatment. Cancer Res . 1993;53:3662-3666. 81. Dedoussis GV, Andrikopoulos NK. Glutathione depletion restores the susceptibility of cisplatinresistant chronic myelogenous leukemia cell lines to Natural Killer cell-mediated cell death via necrosis rather than apoptosis . European Journal of Cell Biology. 2001;80:608-614. 82. Litterst C, Bertolero F, Uozumi J. The role of glutathione and metallothionein in the toxicity and subcellular binding of cisplatin . In: McBrien DCH, Slater TF, eds. Biochemical Mechanisms of Platinum Antitumor Drugs. Oxford : IRL Press; 1986:227-254. 83. Kramer RA , Schuller HM , Smith AC, Boyd MR. Effects of buthionine sulfoximine on the nephrotoxicity of 1-(2-chloroethyl)-3-(trans-4-methylcyclohexyl)-1-nitrosourea (MeCCNU). J Pharmacol Exp Ther. 1985;234:498-506.

63

84. Biaglow JE, Varnes ME, Tuttle SW , et al. The effect of L-buthionine sulfoximine on the aerobic radiation response of A549 human lung carcinoma cells . Int J Radiat Oncol Biol Phys. 1986;12:1139-1142. 85. Ono K, Komuro C, Nishidai T, et al. Radiosensitizing effect of misonidazole in combination with an inhibitor of glutathione synthesis in murine tumors . Int J Radiat Oncol Biol Phys. 1986;12:1661-1666. 86. Shrieve DC, Li GC, Astromoff A , Harris JW . Cellular glutathione, thermal sensitivity, and thermotolerance in Chinese hamster fibroblasts and their heat-resistant variants . Cancer Res . 1986;46:1684-1687. 87. David-Cordonnier MH, Laine W , Joubert A , et al. Covalent binding to glutathione of the DNAalkylating antitumor agent, S23906-1. European Journal of Biochemistry . 2003;270:2848-2859. 88. Inoue H, Takemura H, Kawai Y, Yoshida A , Ueda T, Miyashita T. Dexamethasone-resistant human Pre-B leukemia 697 cell line evolving elevation of intracellular glutathione level: an additional resistance mechanism. Japanese Journal of Cancer Research. 2002;93:582-590. 89. Goto H, Takahashi H, Fuji H, Ikuta K, Yokota S. N-{4-hydroxyphenyl)retinamide (4-HPR) induces leukemia cell death via generation of reactive oxygen species. International J Hematol. 2003;78:219-225. 90. Davison K, Cote S, Mader S, Miller WH. Glutathione depletion overcomes resistance to arsenic trioxide in arsenic-resistant cell lines . Leukemia . 2003; 17:931-940. 91. Kikuchi E, Nakashima J, Horiguchi Y, Oya M , Ohigashi T, Murai M . Enhancement of diethylstilbestrol induced cytotoxicity by bcl-2 antisense oligodeoxynucleotides and a glutathione depletor for prostate cancer. Journal of Urology. 2003;169:730-734. 92. David LL, Calvin HI, Fu S-CJ. Buthionine sulfoximine induced cataracts in mice contain insolubilized crystallins with calpain II cleavage sites . Exp Eye Res . 1994;59:501-504. 93. LaCreta FP, Brennan JM, Hamilton TC, Ozols RF, O'Dwyer PJ. Stereoselective pharmacokinetics of L-buthionine SR-sulfoximine in patients with cancer. Drug Metabolism & Disposition. 1994;22:835-842. 94. Bailey HH, Ripple G, Tutsch KD, et al. Phase I study o f continuous-infusion L-S,R-buthionine sulfoximine with intravenous melphalan. J Natl Cancer Inst. 1997;89:1789-1796. 95. Gallo JM, Brennan J, Hamilton TC, et al. Time-dependent pharmacodynamic models in cancer chemotherapy: population pharmacodynamic mo del for glutathione depletion following modulation by buthionine sulfoximine (BSO) in a Phase I trial of melphalan and BSO. Cancer Res . 1995;55:4507-4511. 96. Skapek SX, VanDellen AF, McMahon DP, et al. Melphalan-induced toxicity in nude mice following pretreatment with buthionine sulfoximine. Cancer Chemother Pharmacol. 1991;28:15-21. 97. Berndt WO. The role of transport in chemical nephrotoxicity. Toxicol Pathol. 1998;26:52-57. 98. Tune BM , Hsu CY, Fravert D. Cephalosporin and carbacephem nephrotoxicity. Roles of tubular cell uptake and acylating potential . Biochem Pharmacol. 1996;51:557-561. 99. Manley HJ, Bailie GR, Eisele G. Bilateral renal cortical necrosis associated with cefuroxime axetil. Clin Nephrol. 1998;49:268-270.

64

100. Andersen JK, Mo JQ, Hom DG, et al. Effect of buthionine sulfoximine, a synthesis inhibitor of the antioxidant glutathione, on the murine nigrostriatal neurons. J Neurochem. 1996;67:2164-2171. 101. White AR, Cappai R. Neurotoxicity from glutathione depletion is dependent on extracellular trace copper. Journal of Neuroscience Research. 2003;71:889-897.

65