Chemical characterization of the degradation products of vinblastine dihydrogen sulfate.

Vinblastine was incubated in 0.2 M glycine buffer (pH 7.4 or 8.8) containing bovine serum albumin (1%) at 37 degrees C for 72 h. The reaction mixture was extracted with CH2Cl2, and the extract gave 6 major peaks (A, B, C, D, E, and F) with retention times of 5.0, 7.5, 11.0, 13.0, 23.0, and 30.0 min, respectively, in a high-performance liquid chromatography system (muBondapak C18 column; solvent, 50% MeOH in 10 mM KH2PO4, pH 4.5; flow rate, 1.2 ml/min; detector, 254 nm). Vinblastine in this system corresponded with peak C, and its spectral data were identical to those of the parent compound. The UV, infrared, and mass spectral properties of these peaks were as follows [UV (lambda max); infrared (cm-1); mass spectrum (m/z)]: peak A: 214, 266, and 315 nm; 3464, 2850, and 1730; and 769 (MH+); peak B: 213, 258, 285, and 295 nm; 3457, 2951, 2580, and 1734; and 809 (MH+); peak C: 214, 266, 292, and 312 nm; 3457, 2951, and 1734; and 811 (MH+); peak D: 212, 266, 285, and 312 nm; 3467, 2915, and 1734; and 811 (MH+); peak E: 212, 260, 285, 294, and 313 nm; 3479, 2850, and 1734; and 825 (MH+); and peak F: 212, 265, 283, and 312 nm; 3407, 2857, and 1734; and 807 (MH+). These data suggest the following tentative structures for the degradation products: peak A, 4-deacetylvinblastine; peak B, 19'-hydroxy-3',4'-dehydrovinblastine; peak D, an isomer of vinblastine; peak E, 19'-oxovinblastine; and peak F, 3',4'-dehydro-19'-oxovinblastine. The structure of peak A as 4-deacetylvinblastine was further confirmed by chemical synthesis.

Structural studies on the degradation products of vincristine dihydrogen sulfate.

Vincristine was incubated at 37 degrees C for 72 h in 0.2 M glycine buffer (pH 7.4 or 8.8) containing 1% bovine serum albumin. The reaction mixture was extracted with CH2Cl2. High-performance liquid chromatography analysis (muBondapak C18 reverse-phase 10-micron steel column; solvent, 50% MeOH in 10 mM KH2PO4, pH 4.5; flow rate, 1.2 ml/min) of the CH2Cl2 extract gave 6 peaks, A, B, C, D, E, and F, with retention times of 4.8, 6.5, 10.0, 12.5, 17.5, and 23.5 min, respectively. Peak C corresponded with vincristine. Spectroscopic data for these peak fractions were as follows [UV (lambda max); infrared (cm-1); mass spectrum (m/z)]: peak A: 220, 256, and 295 nm; 3457, 2922, 1730, and 1669; and 783 (MH+); peak B: 218, 255, and 296 nm; 3435, 2922, 1731, and 1673; and 783 (MH+); peak C: 220, 255, and 296 nm; 3457, 2922, 1738, and 1680; and 825 (MH+); peak D: 218, 252, and 296 nm; 3385, 2922, 1734, and 1677; and 825 (MH+); peak E: 208, 218, 252, and 298 nm; 3371, 2922, 1727, and 1665; and 768 (MH2+); and peak F: 209, 222, 255, and 296 nm; 3392, 2922, 1734, and 1673; and 823 (MH+). These data suggest the following tentative structures for the degradation products: peak A, 4-deacetylvincristine; peak B, an isomer of 4-deacetylvincristine; peak D, an isomer of vincristine; peak E, 4-deacetyl-3-deoxyvincristine; and peak F, N-formylleurosine. The structure of peak A as 4-deacetylvincristine was confirmed by chemical synthesis.

Pharmacokinetics of vincristine, vinblastine, and vindesine in rhesus monkeys.

The pharmacokinetics of vincristine, vindesine, and vinblastine following IV bolus doses of 0.05 mg, 0.10 mg, and 0.20 mg/kg body weight, respectively, were studied in adult male rhesus monkeys. The alkaloid concentrations were determined by a sensitive radioimmunoassay. Pharmacokinetic data were analyzed by a non-linear least-square regression program NONLIN, and the data fit a two-compartment open model. The average terminal half-lives of vincristine, vinblastine, and vindesine in the serum were 189, 152, and 175 min, respectively. The areas under the alkaloid concentration-time curve from 0 to infinity for these drugs for a 1-mg dose were as follows: vincristine, 26,572 nM x min; vinblastine, 16,745 nM x min; and vindesine, 12,708 nM x min. The clearance rate (ml/min/kg) for vincristine (4.8) was slightly lower than that for vinblastine (7.0) or vindesine (7.8). The apparent volumes of distribution for vincristine, vinblastine, and vindesine were, respectively, 1.3, 1.5, and 1.9 l/kg body weight. In the two-compartment open model, the transfer rate constant from compartment 2 to compartment 1 (k21) was lower than the other rate constants (k120 and k12) for each of the alkaloids. The total average excretion of the alkaloids over a 4-day period in urine and feces for vincristine, vinblastine, and vindesine were, respectively, 36.7%, 18.2%, and 25.3% of the injected dose. These data indicate avid tissue retention of the Catharanthus alkaloids in this non-human primate. The similarities between these pharmacokinetic parameters and those previously reported for man suggest that the rhesus monkey is an ideal animal model in which to study the pharmacologic properties of these alkaloids.

Pharmacokinetics of vindesine bolus and infusion.

The pharmacokinetics of vindesine were investigated during treatment of 15 patients with progressive malignancies refractory to conventional treatment. Patients were administered one of three IV dose schedules: 3.0 mg/m2 bolus injection, 1.2 mg/m2/day infusion for 5 days, or 2.0 mg/m2/day infusion for 2 days. Concentrations of the drug in the serum and urine were determined by radioimmunoassay. Serum concentrations were highest (5 X 10(-7) M) in patients receiving a bolus injection, but fell to nondetectable levels by 48 h in four of five patients (terminal t1/2 15.0 +/- 9.4 h). Compared with bolus injection, 1.2- to 1.4-fold greater areas under the blood concentration curve were observed during infusions of 2.0 mg/m2 and 1.2 mg/m2. Whereas steady-state concentrations (approximately 1 X 10(-8) M) were maintained throughout the infusion of 1.2 mg/m2/day progressively increasing serum levels were observed during the infusion of 2.0 mg/m2/day. Serum concentrations fell rapidly following discontinuation of the 2.0-mg/m2 infusion, but were somewhat more sustained in the 1.2-mg/m2 infusion group. The average urinary excretion was similar for each dose-schedule (8%-11% of the total dose). The pharmacokinetics of vindesine are influenced by variations in dose schedule.

Vincristine and vinblastine lower plasma cholesterol concentrations in rhesus monkeys.

3-4 days after a single clinical dose of vincristine or vinblastine in rhesus monkeys there was a marked decrease in plasma low-density lipoprotein cholesterol concentrations. There was also a concomitant increase in plasma triacylglycerol concentrations. Plasma lipid levels returned to normal concentrations within 7-10 days after injection. Plasma high-density lipoprotein cholesterol concentrations were unaltered by the drugs. Electron micrographs of the hepatocytes from monkeys treated with vincristine or vinblastine showed an accumulation of glycogen particles and proliferation of smooth endoplasmic reticulum, which was accompanied by an increase in the number of lipoprotein-containing vesicles. These results indicate that vincristine and vinblastine alter plasma cholesterol and triacylglycerol concentrations in part by interfering with hepatic lipid and lipoprotein metabolism. These studies further suggest the possibility that other less cytotoxic alkaloids from Catharanthus species with clinically useful hypocholesterolemic activity may be discovered.

Continuous intravenous infusion of vinca alkaloid using a subcutaneously implanted pump in a canine model.

A major drawback of infusions of the vinca alkaloids is the lengthy period of hospitalization which is often required for this novel technique of cancer therapy. A potentially useful system to deliver outpatient therapy has been investigated in a preclinical study. A self-contained infusion pump powered by a self-charging fluorocarbon system has been implanted SC in three dogs. The performance of two pumps which had been factory-calibrated to deliver 2.5 and 4.5 ml/day, respectively, was evaluated during 22 infusions of the vinca alkaloids (vincristine, 7; vinblastine, 7; and vindesine, 8). Infusions were given over a 5- to 7-day period and were repeated at 3-week intervals. No malfunctioning of the pumps occurred in over 500 cumulative days of use. The flow rates of the pumps were quite stable except in one animal whose increased flow rate was probably a consequence of fever due to self-induced inflammation about the pump pocket. No local or distant tissue reactions to the pump were observed. Decomposition of vincristine and vinblastine in the infusate at the end of 5- or 7-day infusions was minimal as determined by high-pressure liquid chromatography. The amount of decomposition of vindesine in the infusate was variable. Steady-state concentrations of vincristine during infusion were always greater than 10(-9) M, and were similar to those previously determined in our clinical infusion trials using a dosage of 0.5 mg/m2/day. Clinical evaluation of this system for prolonged infusions of vincristine and other vinca alkaloids appears to be warranted.

Intravenous vincristine infusion: phase I trial.

In an attempt to sustain potentially cytotoxic concentrations of vincristine in man, a five-day continuous infusion of vincristine after an initial intravenous bolus injection was administered to 30 patients with refractory malignancies. Three dosage levels were explored (0.5 mg/m2, 0.75 mg/m2, and 1.0 mg/m2 daily for five days). Neurologic and hematologic toxicity were severe at the high dose level, whereas mild to moderate toxicity occurred at the 0.5 and 0.75 mg/m2 dose levels. Objective responses were noted in 11 patients (37%) with the following malignancies: non-Hodgkin's lymphoma (4), acute non-lymphoblastic leukemia (2), chronic granulocytic leukemia in blast crisis (1), carcinoma of the breast (3), and small cell carcinoma of the lung (1). Responses were observed at each infusion dose level. Nine of the 11 responders had previously progressed while receiving conventional intravenous bolus injection of vincristine. These data suggest that the clinical usefulness of vincristine may be enhanced by the use of infusion techniques. A maximum daily dose of 0.5 mg/m2 given for five days is recommended for future trials of intravenous vincristine infusion.

Pharmacokinetics of vincristine infusion.

Using a sensitive radioimmunoassay, serial blood concentrations of vincristine were measured in 11 patients with refractory malignancies receiving infusions of vincristine at three dose levels: 0.5, 0.75, and 1.0 mg/m2 daily for 5 days. The pharmacokinetics of vincristine infusion were compared to pharmacologic data obtained from four patients who received conventional iv bolus injections of 2 mg of vincristine. Whereas rapid decline of blood levels was seen following iv bolus injection, with concentrations of vincristine approaching 10(-9) M by 48-72 hours, vincristine infusions of 0.5, 0.75, and 1.0 mg/m2 daily for 5 days consistently resulted in blood concentrations greater than 10(-9) M during the treatment period. Similarly, areas under the concentration curve were greater in patients receiving infusions compared to bolus injections of vincristine. Antitumor responses in patients receiving vincristine infusions were observed after failure to respond to iv bolus injections. These data demonstrate the ability of infusion therapy to sustain blood concentrations of vincristine in man beyond that seen with conventional administration and suggest the possibility of improved therapeutic efficacy with this agent by use of infusion techniques.

Pharmacokinetics of vincristine sulfate in adult cancer patients.

Vincristine concentration in serum from 1 min to 72 hr was measured by radioimmunoassay in 14 patients with cancers following i.v. bolus injection of vincristine sulfate at 0.45 to 1.30 mg/sq m. The pharmacokinetic data were analyzed by a nonlinear least-square regression program NONLIN. A three-compartmental open model fitted the raw data better than a two-compartmental model. The mean half-lives of the triphasic decay curves alpha, beta, and gamma were 1.9, 19.2, and 1359 min (22.6 hr), respectively. The apparent volume of the central compartment and the volume of distribution at steady state (Vdss) per 1.73 sq m body surface area were 4.1 and 167.6 liters, respectively. The plasma clearance was 141.9 ml/min/1.73 sq m, and the area under the concentration X time curve from 0 to infinity (AUC0 infinity) for 2 mg vincristine was 21,689 nM.min. Linear regression analysis of the data gave evidence for increasing plasma clearance at higher doses of vincristine. In patients with higher platelet counts, lower AUC0 infinity values were obtained, suggesting a possible interaction of vincristine with blood platelets. Our results, a large Vdss, a long biological half-life, and a low rate-limiting rate constant from Compartment 3 to the central compartment (k31), indicate an avid tissue binding and a slow drug release from the body tissues which may account for drug-related neurotoxicity.

Pharmacokinetics of vincristine in the cerebrospinal fluid of humans.

Using a sensitive radioimmunoassay, concentrations of vincristine and its products were determined in cerebrospinal fluid (CSF) and corresponding blood samples from four patients with lymphoma and two patients with leukemia who had received i.v. bolus injection of 2 mg vincristine. Serial samples of CSF were withdrawn from a CSF reservoir in two patients during a 24-hr period following injection. The first time point after injection at which measurable levels of vincristine were detected in the CSF was 30 min; the concentrations were within the lower range of sensitivity of the assay and were 20- to 30-fold lower than were corresponding serum samples. Despite a prolonged terminal half-life of vincristine in the serum (14.4 to 37.5 hr) following i.v. bolus injection, concentrations of vincristine in the CSF ranged between undetectable within the limits of the assay and 1.1 X 10(-9) M during the 24-hr period of observation. The highest CSF concentration of vincristine (2.6 X 10(-9) M) has been observed in a patient receiving cranial irradiation for active meningeal lymphoma. No measurable levels of vincristine or its products were detected in spot samples of CSF from three patients. Penetration of vincristine and its products into the CSF of humans after i.v. bolus injection appears to be very poor and may account for the uncommon occurrence of central neurotoxicity following its clinical use.

Pharmacokinetics of vincristine sulfate in children.

A radioimmunoassay was used to measure vincristine sulfate concentrations in the serum of four children with malignancies (ages 5-16 years) following intravenous (IV) bolus injections. The pharmacokinetic data were analyzed by a non-linear least-square regression program NONLIN. A three-compartment open model fitted the raw data better than a two-compartment model in three patients. In the other patient the raw data fitted a two-compartment open model. The half-lives of the triphasic decay curves alpha, beta, and gamma were 2.6, 41, and 1,531 min (25.5 h), respectively. The mean apparent volume of the central compartment was 3.25 l, and the the volume of distribution per 1.73 m2 body surface area at steady state was 215.9 l. In a three-compartment open model, the first-order distribution and elimination rate constants (min-1) of vincristine were as follows: k12, 0.088; k13, 0.121; k21, 0.028; k31, 0.0026; k10, 0.045. The plasma clearance was 146.2 ml/min per 1.73 m2, while the AUC0 infinity was 27,816 nM . min. Urinary excretion in one patient demonstrated a drug concentration of greater than 1.0 X 10(-7) M in the urine up to 78 h after the injection. Up to 37% of the administered drug was excreted in the urine as vincristine and/or its metabolites by 90 h. The low elimination rate constant from poorly perfused tissues to blood plasma (k31), a large apparent volume of distribution, and a long biological half-life (25.5 h) indicate avid tissue binding from which a slow release of the drug from the body tissues occurs.

A sensitive radioimmunoassay for vincristine and vinblastine.

A rapid, sensitive and highly reproducible radioimmunoassay (RIA) has been devised for vincristine and vinblastine. These alkaloids bind to the antiserum very tightly with an association constant (Ka) of 5 x 10(9) M-1 as determined by Scatchard analysis. The kinetic association constant of the alkaloids, as determined by the 'on' and 'off' rates at 4 degrees C, is 30-fold higher than the Scatchard analysis value. At 4 degrees C, 80% of the alkaloid is bound to the antiserum within 20 min, and the remaining 20% reaches an equilibrium within 48-72 h. In the RIA procedure the amount of the alkaloid (1-2 ng) needed to compete for 50% binding remains unchanged when the reaction mixtures are incubated for 2, 4, or 20 h. By a modification of the RIA procedure, where the antiserum is first allowed to react with the unlabeled alkaloid followed by addition of the radiolabeled alkaloid, sensitivity of the assay has been increased five- to tenfold. Other drugs, such as adriamycin, CCNU, methyl CCNU, cyclophosphamide, 5-fluorouracil, heparin, hydrocortisone, prednisone, and nitrogen mustard do not interfere with the assay. This assay can thus be used to determine vincristine or vinblastine levels in 5 x 10(-10)--10(-9) M concentrations in biological fluids.

Interaction of 4'-(9-acridinylamino)methanesulfon-m-anisidide with DNA and inhibition of oncornavirus reverse transcriptase and cellular nucleic acid polymerases.

4'-(9-Acridinylamino)methanesulphon-m-anisidide (AMSA) (NSC 141549), an acridine derivative with activity against a variety of laboratory tumors in vivo, is presently undergoing Phase 1 clinical evaluation. The interaction of AMSA with DNA and its effects on nucleic acid-polymerizing enzymes were examined in an attempt to define the site of cytotoxicity of AMSA. Binding of AMSA to DNA, as demonstrated by equilibrium dialysis and spectrophotometric methods, appears to be similar to other aminoacridines, in that two types of binding sites (type 1 and type 2) were observed. Fluorescence studies and thermal denaturation studies gave strong evidence that AMSA type 1 binding was by intercalation into DNA. The binding of AMSA to DNA was without marked base-pair specificity. Furthermore, the effect of AMSA on nucleic acid-polymerizing enzyme activities (mouse embryo DNA polymerase alpha, avian myeloblastosis virus reverse transcriptase, and Escherichia coli RNA polymerase) was studied. Inhibition of enzyme activity by AMSA appeared to be independent of DNA base sequence. The relatively high concentrations of AMSA required for inhibition of these enzymes as compared to the concentrations of AMSA necessary for cytotoxicity in vitro suggest that the interaction with DNA alone might not fully explain its antitumor activity.

Base specificity in the inhibition of oncornavirus reverse transcriptase and cellular nucleic acid polymerases by antitumor drugs.

Adriamycin, daunomycin, acridylmethanesulfonanilide, and alkoxybenzophenanthridine alkaloids (coralyne acetosulfate, fagaronine chloride, and nitidine chloride) inhibit template-directed nucleic acid polymerizing enzyme activities like reverse transcriptase, DNA polymerase, and RNA polymerase. Enzyme reactions with poly(dA-dT), poly(rA)-oligo(dT) and poly(dA)-oligo(dT) are more strongly inhibited by the drugs than those with poly(dC)-poly(dG) and poly(rC)-oligo(dG). These results suggest that the antitumor drugs inhibit nucleic acid polymerases by a specific interaction with A:T base pairs of the templates.

Interaction of rifamycins with mammalian nucleic acid polymerizing enzymes.

Procedures were established for the isolation and partial purification of DNA polymerase, RNA polymerase and poly(A) polymerase activities from the cytoplasm and nuclei of NIH-Swiss mouse embryos. Based on the elution pattern of these enzyme activities from DEAE-cellulose and phosphocellulose columns in Tris-HCl buffer, pH 8.0, the apparent basicities of the enzymes can be arranged as follows: cytoplasmic(C) poly(A) polymerase greater than (C)DNA polymerase beta greater than (C)DNA polymerase alpha and nuclear(N) poly(A) polymerase greater than (N)DNA polymerase greater than (N)RNA polymerase I greater than (N)RNA polymerase II. Twenty rifamycins, including rifamycin B, rifamycin S, rifamycin SV, and rifamycin SV derivatives, were examined for their ability to inhibit the above mentioned nucleic acid polymerizing enzymes and Simian sarcoma virus type I (SSV-1) reverse transcriptase. Rifamycin SV 3'-formyldiphenylhydrazone, rifamycin SV 3'-formyl-n-octyloxime (AF/013) and rifamycin SV 3'-formyldiphenylmethyloxime (AF/05) inhibited all the tested enzyme activities. Rifamycin SV 3'-formylpropylphenyloxime (AF/015) inhibited cellular nucleic acid polymerase activities but not SSV-1 DNA polymerase activity. Rifamycin SV 3'-formyldinitrophenylhydrazone (AF/DNFL) strongly inhibited reverse transcriptase activity but did not inhibit cellular DNA polymerase activities. AF/DNFI slightly inhibited RNA and poly(A) polymerase activities. Rifamycin SV 3'-formyldipropylhydrazone (AF/DPI) and 2,6-dimethyl-4-N-benzyldemethyl-rifampicin (AF/ABDMP) slightly inhibited reverse transcriptase activity but did not inhibit cellular nucleic acid polymerase activities. Active rifamycin derivatives inhibited enzyme reactions by interacting with the enzyme proteins. Nascent polynucleotide chain elongation continued although at a reduced rate in the presence of inhibitor. The addition of increasing concentrations of nonionic detergent (Triton X-100) to rifamycin-inhibited enzyme reactions fully restored enzyme activities. The presence of highly lipophilic 3'-side chains on active rifamycins and the reversibility of enzyme inhibition by Triton X-100 suggest that the tested nucleic acid polymerizing enzymes may have hydrophobic regions with which inhibitory rifamycins interact.

Inhibition of mammalian and oncornavirus nucleic acid polymerase activities by alkoxybenzophenanthridine alkaloids.

The alkoxybenzophenanthridine alkaloids (coralyne acetosulfate, fagaronine chloride, and nitidine chloride) have been reported to possess antileukemic activity in mice. These compounds were tested for inhibition of reverse transcriptase activity of an RNA tumor virus and DNA polymerase, RNA polymerase, and polyadenylic acid polymerase activities of NIH-Swiss mouse embryos. Reverse transcriptase and DNA polymerase activities were strongly inhibited by these antileukemic alkaloids, whereas RNA polymerase and polyadenylic acid polymerase activities were only moderately affected. Viral and cellular DNA polymerase activities were potently diminished by the alkaloids when poly[d(A-T)], poly(dA)-oligo(dT), and poly(rA)-oligo(dT) template primers were used in the reaction mixture; however, no inhibition of enzyme activity was obtained with poly(rC)-oligo(dG) as template primer. These results suggest that alkoxybenzophenanthridine alkaloids inhibit DNA polymerase activity by interaction with A:T base pairs of the template primer.