Background: 4-Hydropeoxyifosfamide (HOOI) is a hydroperoxy derivative of ifosfamide that was developed as an anticancer agent that can penetrate the blood-brain barrier (BBB), which can be potentially useful in the management of brain tumors.

all groups, the control group, no dose-dependent the attributable 0.3% Klucel+1.92% Tween 80 vehicle rather than 1 . Pertinent clinical signs emesis, righting function skin discoloration (entire body, ears, or face), swelling (face and/or nose/muzzle), eyes slow gum capillary refill time, feces-mucoid/soft/discolored/ lacrimation, salivation, sclera injected, vocalization, tremors, and urination decreased. report here the results of acute toxicity and pharmacology studies with single intravenous of 1 in groups of and dogs. The end-point of all the studies to identify and an acceptable starting dose for a Phase I clinical trial in humans with

e13523 Background: Recent studies have shown that siRNA knockdown of the cGMP-specific phosphodiesterase, PDE5 can suppress growth and induce apoptosis of human breast tumor cells. However, conventional PDE5 inhibitors have only modest in vitro tumor cell growth inhibitory activity at concentrations that cannot be achieved in vivo. We hypothesized that this may be attributed to the reversible nature of their binding to PDE5. Methods: A series of alkyl halide derivatives of the PDE5 inhibitor, tadalafil were synthesized and evaluated for in vitro tumor cell growth inhibitory activity, mechanism of action, tumor selectively, and in vivo antitumor activity. Results: Among 18 derivatives synthesized, 15 inhibited the growth of human breast MDA-MB-231 tumor cells with IC50 values in the 10-9 to 10-6 M range. Compound 6 potently inhibited tumor cell growth with an IC50 of 6 nM and induced apoptosis. Normal human mammary epithelial cells that do not express PDE5 were insensitive to treatment, while breast tumor cell lines that express PDE5 were highly sensitive. Compound 6 selectively suppressed cGMP hydrolysis in lysates from treated cells and increased intracellular cGMP levels. Moreover, intermittent treatment (3hrs) suppressed growth as effectively as continuous treatment (72hrs). These observations along with molecular docking studies are consistent with a mechanism involving irreversible PDE5 inhibition. Screening in the NCI-60 panel revealed striking sensitivity among most human tumor cell lines from breast, renal, CNS, ovarian, and hematological origin with IC50 values in the low nanomolar range. Compound 10 was predicted to have favorable pharmacokinetic properties and evaluated for in vivo antitumor efficacy. With an IC50 of 2 nM to inhibit human Caki-1 renal tumor cell growth, compound 10 significantly suppressed tumor growth in the Caki-1 xenograft mouse model. Conclusions: Mono-alkylating derivatives of tadalafil have potential therapeutic utility for a broad range of solid and hematological tumors.

Abstract HOOI is a hydroperoxy-derivative of ifosfamide (IFOS) and a pro-drug of isophosphoramide mustard (IPM) [the active metabolite of IFOS] – a bi-functional alkylator that cross-links with G/C DNA base sequences resulting in irreparable inter-strand DNA cross-linking and cell death. HOOI has been an unstable laboratory curiosity for years; however, HOOI.L-lysine is a stable salt complex that has allowed the development of HOOI for clinical trials. HOOI spontaneously releases acrolein and chloroacetaldehyde in situ in cancer cells not extracellular in the general circulation (as does IFOS) and has not been associated with cystitis, renal tubular necrosis and/or CNS toxicity. HOOI has been screened in 20+ human xenograft tumor and murine tumor models and has significantly improved %ILS in intracranially implanted human xenografts – MX-1, U251; ZR-75-1 and in P399/CPA leukemia vs. IFOS and IPM. The U251 data (%ILS – 84+) is impressive considering – BCNU produced a 72% ILS. The ZR-75-1 breast cancer had a %ILS – 83. Bone marrow failure was the DLT in mice – LD10 100 mg/kg (for M/F). Dogs (M/F) were treated with 10, 15, 20 and 30 mg/kg. For both sexes, the LD10 was calculated to be 17.2, while the LD50 was calculated to be 17.3 mg/kg. PK values in dogs revealed the following profile for groups dosed with 30 mg/kg: AUC0- t = 1.53 (mg h/L), T1/2α = 0.93 (h), T1/2β = 6.1 (h) & CL = 19.5 (L/ h) [a two compartment model]. The AUC was linear for the 10 and 30 mg/kg doses. HOOI did not generate any detectable plasma chloroacetaldehyde vs. IFOS (2.12 µg/mL from 400 mg/kg) and 25% of the acrolein generated by the MTD of CPA & IFOS. No convulsions, neuropathies or renal dysfunction were observed in either specie. Unlike IFOS/IPM, HOOI is lipophilic, activated intracellular (with < extracellular acrolein and no chloroacetaldehyde released); no IFOS/IPM-associated CNS or GU toxicity was noted. HOOI may increase the safety and efficacy margins of this class of alkylators in advanced CPA- and IFOS-resistant cancers and broaden the target-range (CNS-gliomas). The L-lysine salt stability supports developing HOOI for clinical trials. Supported by grant R44 CA094566 from the NCI/SBIR program. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 102nd Annual Meeting of the American Association for Cancer Research; 2011 Apr 2-6; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2011;71(8 Suppl):Abstract nr 3222. doi:10.1158/1538-7445.AM2011-3222

Abstract Introduction: 4-Demethyl-4-cholesteryloxycarbonyl-penclomedine (DM-CHOC-PEN) is a polychlorinated pyridine cholesteryl carbonate, which is in Phase I clinical trials in patients with advanced cancer - IND 68,876. DM-CHOC-PEN is active vs. intracranially (IC) implanted human xenograft models - U251 and D54 glioblastoma and MX-1 breast cancer and recently found to be active vs. B-16 melanoma - 142% ILS. DM-CHOC-PEN's MOA is via alkylation of DNA @ N7 - guanine, as well as converting melanoma cells into a melanotic Go phase with cell death. Not all cells were converted into Go phase and senescence, thus the interest in designing a binary drug approach for DM-CHOC-PEN, as an improvement in treatment for melanoma. A number of agents are cytotoxic vs. B-16 cells in vitro and in vivo; however, 4-hydroperoxyifosfamide (HOOI), which is converted to isophosphoramide mustard (IPM) in vivo, demonstrated the best %ILS. The latter drug has been chosen as a companion with DM-CHOC-PEN in the present binary drug study vs. B-16 mouse melanoma. Methods: B-16 melanoma cells were cultured using RPMI media with 10% FBS and pen/strep @ 37° C in a CO2 incubator. Drugs were added to the cells in a growth phase and removed after 8–12 h. Adult female C57BL mice in groups of 5–6 mice were implanted subcutaneously (SC) with B-16 mouse melanoma (106 cells) and when SC nodules were palpable the mice were dosed IP daily (200 mg/kg) for 5-days with DM-CHOC-PEN followed by HOOI administered IP @ varying doses and daily/weekly schedules and monitored daily until death. Mice with SC B-16 melanoma were dosed with single agents - DM-CHOC-PEN, HOOI, cis-platinum and temozolamide, which were used as controls. Tumor tissue was extracted with dichloromethane, and assayed per HPLC and NMR. Results: In vitro, DM-CHOC-PEN and HOOI, as single agents, had IC50 of 0.5 and 0.8 μg/mL vs. B-16 melanoma cells, resp. Mice bearing SC B-16 melanoma treated with DM-CHOC-PEN (200 mg/kg/d × 5d, IP) alone vs. saline controls demonstrated %ILS of 142%; thus supporting the in vitro observations. In vivo in the B-16 melanoma model, cis-platinum as a single agent had a %ILS = 0%, but for HOOI it was 85%. In 2-drug studies, DM-CHOC-PEN plus cis-platinum together (in theoretical therapeutic ranges) were too toxic in combination, however, the %ILS for DM-CHOC-PEN plus HOOI was 173%. Tumor tissue was removed within 2-days of treating mice with DM-CHOC-PEN, extracted and revealed 75 μg/g tumor tissue of DM-PEN, a metabolite. Discussion: HOOI is a S-phase alkylating agent that is an appropriate 2nd agent to kill cells escaping from the DM-CHOC-PEN - induced Go phase induction. To date, the best treatment regimen was DM-CHOC-PEN (200 mg/kg/d × 5d) followed by HOOI (90 mg/d × 3d) - %ILS = 173. The finding that melanoma tissue extracts resulted in μg of DM-PEN (the metabolite) is in agreement with the pharmacokinetic findings observed for DM-CHOC-PEN in rats and humans, which will be reviewed. The binary drug combination will be reviewed with the FDA. Supported in part by: NCI SBIR grants - R43/44CA85021. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the AACR-NCI-EORTC International Conference: Molecular Targets and Cancer Therapeutics; 2011 Nov 12-16; San Francisco, CA. Philadelphia (PA): AACR; Mol Cancer Ther 2011;10(11 Suppl):Abstract nr B219.

Current information indicates that asparagine-linked oligosaccharides of glycoproteins arise from a precursor whose probable composition is dolichol-P2-(N-acetylglucosamine)2(mannose)9(glucose)3. This compound is formed by the stepwise addition of saccharide units to dolichol phosphate, presumably yielding 14 components. A procedure is presented in this chapter that rapidly resolves these 14 dolichylpyrophosphoryl oligosaccharides by liquid chromatography on silica gel. Liquid chromatography of dolichol-linked oligosaccharides is carried out on six 0.32 × 100 cm columns linked in series, packed with silicic acid and maintained at 58 °. A 0.32 × 3.5 cm precolumn with the same packing can be used. Separation of the dolichylpyrophosphoryl oligosaccharides is carried out with a nonlinear gradient formed with solvent A, CHCl3-CH3OH-conc. NH3 and solvent B, CHCl3-CH3OH conc. NH3-H2O. Both solvents contain 0.6 g of ammonium chloride per liter. The gradient is run for 120 min from 0 to 100% solvent B, followed by pumping solvent B for 75 min. The solvent composition during the gradient at t minutes is given by: % solvent B = 100(t/120).

Purpose The purpose of this investigation was to synthesize a series of carbonate and carbamate derivatives of 4-demethylpenclomedine (DM-PEN), the major plasma non-toxic metabolite of penclomedine (PEN) seen in patients. DM-PEN has been observed to be an active antitumor agent in mouse human xenograft tumor models and non-neurotoxic in a rat model, however, activity in intracranially implanted human glioma xenograft models have not been reported. The major goal was to identify derivatives that are active in brain tumors. Methods Derivatives were prepared from DM-PEN and evaluated in vivo against human U251 glioblastoma, D54 glioblastoma and MX-1 breast tumor xenografts and mammary tumor 16/C that were implanted in the mammary fat pad or intracranially (IC). Results Carbonate and carbamate derivatives were found to be superior to DM-PEN against IC growing human glioblastoma xenografts. Conclusion The activity of the carbonates and carbamates against human tumor xenografts in vivo suggests consideration of these two series of derivatives of DM-PEN for clinical development.

Treatment of cancer patients with chemotherapeutic drugs is often associated with the occurrence of tumors with a multidrug resistance (MDR). Furthermore, the relation between overexpression of P-glycoprotein (P-gp) and resistant cancers has been well established. In this study, novel 2-aminobenzofuran derivatives were synthesized and tested for their ability to modulate P-gp mediated multidrug resistance (MDR) in vitro. The most potent compound, 43, increased P-gp inhibitory activity at 5 μM by 11.12-fold and was 3.6-fold stronger than verapamil. Furthermore, 43 can sensitize Flp-In™-293/MDR cells toward vincristine, paclitaxel and doxorubicin by 17.95-fold, 13.68-fold and 26.43-fold at 2.5 μM, respectively. 43 also can sensitize the resistant cancer cell line KBvin toward vincristine, paclitaxel and doxorubicin by 246.43-fold, 38.72-fold and 5.16-fold at 2.5 μM, respectively. In conclusion, important aspects for developing potent P-gp inhibitors have been emphasized in this study, providing a starting point for the further structural optimization of P-gp inhibitors.

A280 4-DM-PEN is a non-neurotoxic metabolite of penclomedine (PEN) with anti-cancer properties. However, 4-DM-PEN does not readily cross the blood brain barrier and does not produce complete remissions in intracerebral (IC) gliomas. PEN did produce responses in clinical trials (gliomas - CR), but was neurotoxic and all trials ceased. DEKK-TEC and SRI have synthesized 20 carbonate and carbamate derivatives of 4-DM-PEN. Analogs prepared were from the series: 4-DM-PEN-4-OCO2-X & 4-DM-PEN-4-OCONH-X, where X = benzyl, methyl, ethyl, octyl, 4-Cl-, 4-F-, 4- & 2-nitrobenzyl, phenyl, 4-Cl-, 4-F-, 4-nitrophenyl, N-morpholino and cholesteryl groups. Anti-cancer activities were noted with both groups when administered IP daily x 5 days to SC growing MX-1 xenografts [response ranges - %ILS >50% and 20-40% CR]. However, only a carbonate, 4-demethyl-4-cholesteryloxy-penclomedine (DM-CHOC-PEN, X=cholesteryl) was active (produced CRs) vs. three IC implanted xenograft tumor (U251, D54 & MX-1) models with no weight loss. BCNU controls did not produce CRs in IC implanted glioma xenografts, such as D-54 in mice. The IC tumor responses for DM-CHOC-PEN are the platform for our interest in evaluating the latter drug as clinical treatment for 1o and 2o CNS malignancies in humans. The IC activity is in contrast to DM-PEN, which was active vs. MX-1 breast tumor xenografts growing SC in mice but did not produce CRs in IC implanted human U-251and D-54 glioblastoma multiforme xenografts and MX-1 xenograft models. DM-CHOC-PEN vs DM-PEN has improved activity (% ILS/CR) in IC implanted human xenograft models - U251 glioma: +29/25 vs 17/0, resp. and MX-1 breast cancer: +20/17 vs 12/0, resp. DM-CHOC-PEN’s acute toxicology in mice and dogs has been reported - AACR 48, abst. 5614, 2007. Mechanisms for CNS anticancer activity and clinical plans for Phase 1 trials will be discussed. Supported by NCI/SBIR grant - 5R44CA85021.

9524 Background: IPM is a bi-functional alkylator which cross-links DNA through G:C base-pairs resulting in irreparable 7-atom inter-strand cross-links. IPM is the active moiety of ifosfamide (IFOS), a pro-drug of IPM. IPM is active in diverse cancer models but is unstable. We stabilized IPM with lysine (IPM-lysine; ZIO-201). ZIO-201 was active in pre-clinical models including human cancer cell lines, human-mouse xenografts and cancers resistant to cyclophosphamide (CPA) and IFOS. Because ZIO-201 is not metabolized to acrolein or chloroacetaldehyde, bladder and CNS toxicities are unlikely. Methods: Phase-1 trial in subjects with advanced cancers. ZIO-201 was given daily for 3 consecutive d at a starting dose of 30 mg/me2/d every 3 w. Neither mesna nor IV hydration were given. 11 dose levels were studied in 18 subjects up to 795 mg/me2/d; dose-escalation continues. Data on the 1st 15 subjects are available for analysis. Results: Median age was 59 y (range, 18–70 y); 10 subjects were male. Diagnoses included colorectal cancer (N=5), sarcoma (N=3) and 1 subject each with gastric, lung, bladder, prostate, ovary and thyroid cancers and mesothelioma. 7 had extensive and 8, limited disease. All subjects received extensive prior therapy. Median N cycles was 2 (range, 1–13). Toxicities ≥ grade-2 occurring in > 20% of subjects included anemia (N subjects=4) and diverse GI complaints (N=4). 4 of 8 subjects receiving doses > 445 mg/me2/d had transient proximal renal tubular acidosis. There was no hemorrhagic cystitis or CNS toxicity. 1 subject with mesothelioma had stable disease > 13 mo. Pharmacokinetic studies at 595 mg/me2/d showed a tmax = 13 min (SD ± 9 min), Cmax = 44.7 μg/mL (SD ± 34.1 μg/mL), t1/2 = 35 min (SD ± 7 min) and AUC0-∞ = 1.68 mg·min/ml (SD ± 1.26 mg·min/ml). Conclusions: These data suggest a possible role for ZIO-201 in IFOS-sensitive cancers (especially sarcoma and lymphoma). ZIO-201 may also be active in CPA and IFOS-resistant cancers. Comparable or greater efficacy with less toxicity is expected. [Table: see text]

The hepatic cytochrome P-450-mediated metabolism and met abolic activation of [cWoroef/7y/-3H]cyclophosphamide ([cWoroetfjy/-3H]CP) and [4-14C]cyclophosphamide ([4-14C]CP) were in vestigated in vitro in the reconstituted system containing cyto chrome P-450 isolated from phenobarbital-treated rats. In addi tion, hepatic microsomal binding and the hepatic microsomemediated metabolism of [14C]acrolein, a metabolite of [4-14C]CP, were also investigated. The metabolism of [c/J/oroef/7y/-3H]CP and [4-14C]CP to polar metabolites was found to depend on the presence of NADPH and showed concentration dependence with respect to cyto chrome P-450 and NADPH:cytochrome P-450 reducÃ-ase.K„ and Vmaxvalues were essentially similar (Km,0.44 and 0.42 mW; V™», 4.8 and 7.0 nmol of polar metabolites formed/min/nmol of cyto chrome P-450 for [4-14C]CP and [cWoroef/iy/-3H]CP, respec tively). The patterns of inhibition by microsomal mixed-function oxidase inhibitors, anti-cytochrome P-450 antibody, and heat denaturation of the cytochrome P-450 were essentially similar, with subtle differences between [4-14C]CP and [c/7/oroef/)y/-3H]CP metabolism. The order of inhibition by various mixed-function oxidase inhibitors was SKF > aand 0-naphthoflavones > metyrapone. The in vitro metabolic activation of CP in the reconstituted system demonstrated predominant binding of [cWoroef/7y/-3H]CP to nucleic acids and almost exclusive binding of [4-14C]CP to proteins. Gel electrophoresis-fluorography of the proteins in the reconstituted system treated with [4-14C]CP demonstrated lo calization of the 14C label in the cytochrome P-450 region. To examine this association further, hepatic microsomes were mod ified with [14C]acrolein in the presence and the absence of NADPH. The results confirmed covalent association between [14C]acrolein and cytochrome P-450 in the microsomes and also demonstrated further metabolism of [14C]acrolein, apparently to an epoxide, which is capable of binding covalently to proteins. The results of these investigations not only confirm the signif icance of primary metabolism but also emphasize the potential role of the secondary metabolism of cyclophosphamide in some of its toxic manifestations.

PURPOSE:The purpose of this investigation was to synthesize a series of thiolo-, thiono- and dithiocarbonate and thiocarbamate derivatives of 4-demethylpenclomedine (DM-PEN), the major plasma metabolite of penclomedine (PEN) in patients observed subsequently to be an active antitumor agent and non-neurotoxic in a rat model, in order to compare their antitumor activity with that of DM-PEN.METHODS:Derivatives were prepared from DM-PEN and evaluated in vivo against human MX-1 breast tumor xenografts implanted in the mammary fat pad, several of which were also evaluated against human brain tumor xenografts.RESULTS:Thiolocarbonate and thiocarbamate derivatives were found to be superior to DM-PEN against MX-1 tumor and modestly active against glioblastoma.CONCLUSION:The activity of the thiolocarbonates and thiocarbamates against human tumor xenografts in vivo suggests consideration of these two series of derivatives of DM-PEN for clinical development.

Ifosfamide (IFOS) and cyclophosphamide (CPA) are widely used anti-cancer drugs. Both are pro-drugs which must be metabolized before their metabolites cross-link DNA. Use of these drugs, especially high-dose IFOS, is complicated by substantial adverse effects of metabolites not directly involved in DNA-cross-linking including kidney, bladder and CNS toxicities. Recently, a lysine-stabilized form of isophosphoramide mustard (IPM-lysine; ZIO-201), an IFOS-metabolite, was developed. ZIO-201 directly cross-links DNA and is active against human cancer cell lines and in mice with human cancer xenografts. To evaluate nephrotoxicity of ZIO-201 we used an in vitro primary rabbit kidney proximal tubule (RPT) cell culture system, which retains many of the characteristics of renal proximal tubule cells including a polarized morphology, a Na+/glucose cotransport system and a p-aminohippurate transport system, glutathione status, and hormone responses (including a parathyroid hormone sensitive adenylate cyclase). We studied effects of ZIO-201 and other IFOS metabolites on primary RPT cells. IFOS is activated by the cytochrome P450 system producing a mixture of 2- and 3-dechloroethylifosfamide and resulting in formation of chloroacetaldehyde (CAA). Also, metabolism of IFOS by the cytochrome P450 system produces 4-hydroxyifosfamide, ultimately resulting in the production of acrolein (ACR). CAA and ACR are IFOS metabolites implicated in kidney toxicity in humans. Effects of ZIO-201 and CAA on the viability of confluent monolayers of primary rabbit kidney proximal tubule (RPT) cells were tested at 15–100 μM. Viability was determined by neutral red dye uptake by the primary cultures after 8 h incubation. ZIO-201 did not significantly reduce viability of primary RPT cells; the LD50 for CAA was about 40 μM with complete inhibition at 75 μM. Effects of 4-hydroperoxyifosfamide (HIFA), a synthetic prodrug of 4-hydroxyifosfamide and ACR were also studied. Although ACR had an LD50 of 80 μM HIFA did not significantly affect viability. The IFOS metabolite 4-hydroxyifosfamide, was prepared from 4-hydroperoxyifosfamide (1). At 200 μM, 4-hydroxyifosfamide resulted in extensive cell death after 4 h incubation. These data show that ZIO-201 avoids the kidney toxicity caused by metabolites of pro-drugs like IFOS. Whereas 2 IFOS metabolites, CAA and ACR were cytotoxic to kidney cells, ZIO-201 was not. An explanation for the toxic effects of 200 μM 4-hydroxyifosfamide t might be metabolism to ACR in vitro.

Isophosphoramide mustard (IPM) is known to have substantial anti-cancer activities in various animal models. Liquid chromatography–electrospray mass spectrometry (LC–ES–MS) and LC–ES–MS/MS methodologies have been developed and applied to the analysis of synthesized preparations of IPM. Our studies reveal that the principal impurity in IPM is N-(2-chloroethyl)-N′-ethylphosphorodiamidic acid (MC-IPM) formed by dehydrochlorination of IPM with subsequent hydrogenation during synthesis. This impurity is present at levels in the range of 2–5% depending upon synthesis conditions. In addition, a second IPM derivative has been characterized by LC–ES–MS/MS and has been shown to be the product of a reaction of IPM with the dilute perchloric acid mobile phase used for liquid chromatography separations. The LC–ES–MS/MS method has been successfully employed to detect IPM spiked into a blood plasma sample. This work establishes that LC–ES–MS/MS is a viable tool for the detailed characterization of IPM and related products.

Background Isophosphoramide mustard (IPM) is the cytotoxic alkylating metabolite of Ifosfamide (IFOS). IPM is being readied for a phase I clinical trial. In the present preclinical study, IPM was evaluated for usage in multidose intravenous (IV) infusion protocols. Methods Mice and dogs received IV IPM daily for 3 days. Single-day dosing—oral and IV—to mice, rats, and monkeys is also reviewed for comparison. Complete toxicology studies were completed in the mice and dogs. For mice, dogs and monkeys, IV pharmacokinetic studies were conducted and compared. Results For mice, the LD 10 for the 3-day IV schedule for IPM was calculated to be 119 mg/kg (with 95% confidence limits of 87–134 mg/kg) (combined sexes), and for adult male dogs the maximum tolerated dose (MTD) was 5 mg/kg. Pharmacokinetic studies in mice, dogs and monkeys were compared and projected to human dosing. For dogs that received 10 mg/kg of IPM, T 1/2β was 0.99 h, and clearance was constant (1.01 l/h/kg). IPM was detected from 0 h to 1.5 h after the 5 mg/kg dose and from 0 h to 2 h after the 10 mg/kg dose; none was detected after 2 h. The IV MTD in dogs was 5 mg/kg per day for 3 days. Renal tubular necrosis and bone marrow failure were the causes of death. Transient liver, renal and bone marrow toxicity and gastrointestinal dysfunction were seen at low doses (<5 mg/kg) in dogs. In mice (receiving 100 mg/kg IV) plasma concentrations disappeared in less than 1 h (T 1/2α 2 min), with a clearance of 8.44 l/h/kg. For monkeys, the mean T 1/2 was 4.2 h. Median clearance was 1.65 l/h/kg and no IPM was detected 4 h after dosing. No potential IPM metabolites could be detected in any of the studies. In vitro, plasma protein bound 90% of IPM within 5 min of incubation. Conclusions Predictions for human pharmacokinetic parameters and dosing are made from allometric analysis using the above three species. Data predicted an acceptable starting dose of 30 mg/m 2 with a clearance of 39.5 l/h, and a T 1/2 of 1 h 45 min for a 70-kg patient.

Purpose. Penclomedine (PEN), a multichlorinated α-picoline derivative which is metabolized to highly reactive alkylating species, was selected for clinical development due to its prominent activity against a wide range of human tumor xenografts when administered either parentally or orally. Its principal dose-limiting toxicity in preclinical and clinical studies has been neurocerebellar toxicity, which has been related to the magnitude of peak plasma PEN concentrations, but not to plasma concentrations of its putative principal alkylating metabolite, 4,o-demethylpenclomedine (DMPEN). These observation, as well as PEN's toxicologic, pharmacologic, and tissue distribution profiles, have suggested that the parent compound is primarily responsible for cerebellar toxicity. The studies described in this report were undertaken to characterize the neuropathology of PEN neurotoxicity, with a long-term goal of developing strategies to maximize its therapeutic index. Design. Male Sprague–Dawley rats were treated with therapeutically relevant doses of PEN, orally and intraperitoneally (i.p.), on various administration schedules, and DMPEN administered i.p. The animals were monitored for neurotoxicity, and brain sections were examined for neuropathology, particularly Purkinje cell loss and neuronal injury. Brain sections were stained using standard histochemical techniques and immunostained with OX-42 to detect microglial cells that are activated following neuronal damage, and calbindin D28K, a calcium-binding protein expressed by cerebellar Purkinje cells. Results. Dose-related neurocerebellar toxicity associated with parasagittal bands of Purkinje cell degeneration and microglial activation in the cerebellar vermis were evident in rats treated with PEN 100–400 mg/kg i.p. as a single dose. Neuronal injury was not observed in other regions of the brain. Furthermore, neither clinical nor histopathological evidence of cerebellar toxicity was apparent in rats treated with similar total doses of PEN administered i.p. on a daily×5-day dosing schedule. Similar histological findings, in an identical neuroanatomical distribution, were observed in rats treated with PEN orally; however, the magnitude of the neuronal toxicity was much less than in animals treated with equivalent doses of PEN administered i.p. Although acute lethality occurred in some rats treated with equimolar doses of DMPEN as a single i.p. treatment, surviving animals exhibited neither signs nor histopathological evidence of neurocerebellar toxicity. Conclusions. PEN produces selective dose- and schedule-dependent Purkinje cell degeneration in the cerebellar vermis of rats, whereas therapeutically relevant doses of PEN administered orally are better tolerated and produce less neurocerebellar toxicity. In addition, roughly equivalent, albeit intolerable, doses of the major active metabolite DMPEN, which was lethal to some animals, produced neither clinical manifestations of neurocerebellar toxicity nor Purkinje cell loss. These results support a rationale for investigating whether PEN administered orally, which may undergo significant first-pass metabolism to DMPEN and other less toxic intermediates, or treatment with DMPEN, itself, may result in less neurocerebellar toxicity and superior therapeutic indices than PEN administered parenterally.