Effect of inline filtration on ViaSpan cold-storage solution.

The effect of filtration on the particulate load in ViaSpan cold-storage solution was studied. Commercially available inline blood transfusion filters (SQ40S, Pall Biomedical) were inserted into the delivery-set port of polyvinyl chloride bags of ViaSpan (DuPont Merck Pharmaceutical). Particles in samples collected in particle-free vials before and after filtration were counted by a light-obscuration technique. The compatibility of key ViaSpan ingredients and of three commonly used additives (penicillin G potassium, dexamethasone sodium phosphate, and human insulin) with the inline filter was also investigated. Filtration removed 56% of particles 5-10 microm in diameter, 71% of particles of > 10-25 microm, and > 99% of particles measuring > 25 microm. Flow rates with the filters were more than adequate for clinical use. Concentrations of constituent hydroxyethyl starch, glutathione, adenosine, allopurinol, raffinose pentahydrate, and lactobionic acid were not significantly affected by filtration, nor were the concentrations of any of the drug additives. Inline filtration of ViaSpan effectively reduced fatty acid salt particulates in the solution and had no deleterious effect on flow rate, ingredient concentrations, or concentrations of commonly added drugs.

A comparison of oral and rectal absorption of L-dopa esters in rats and mice.

Short-chain alkyl esters of L-dopa were administered to rats and mice via oral and rectal routes. Plasma L-dopa esters and L-dopa were determined in the systemic and portal circulation by HPLC. A comparison of isopropyl, butyl, and 4-hydroxybutyl esters of L-dopa demonstrated significantly higher levels of the esters in both systemic and portal blood samples following rectal administration than following oral administration. In most cases, oral administration resulted in nondetectable (less than 0.01 micrograms/ml) levels of the esters in plasma. Correspondingly, the plasma levels of L-dopa itself were consistently higher following rectal administration. At very high oral doses (500 mg L-dopa equivalents/kg body weight), systemic plasma levels of the butyl ester could be detected (1.25 micrograms/ml at 10 min), which might indicate saturation of the esterase activity of the small intestine. These studies indicate that the systemic availability of L-dopa from short-chain alkyl esters of L-dopa may be best optimized by rectal administration, which avoids the relatively high esterase activity characteristic of the small intestine.

Buccal absorption. III. Simultaneous diffusion and metabolism of an aminopeptidase substrate in the hamster cheek pouch.

The simultaneous diffusion and metabolism of the D- and L-isomers of the aminopeptidase substrate, leucine-p-nitroanilide (LPNA), were examined in vitro in the hamster cheek pouch. L-LPNA was completely hydrolyzed during diffusion across the cheek pouch, whereas D-LPNA crossed the cheek pouch intact. The metabolic barrier appeared to be localized in the epithelium of the cheek pouch. Addition of an aminopeptidase inhibitor, bestatin, to both diffusion cell reservoirs resulted in decreased hydrolysis of L-LPNA. The experimental results were analyzed with a mathematical model which was developed to describe the simultaneous diffusion and metabolism processes. Using this model it was estimated that the rate of diffusion of L-LPNA across the cheek pouch was less than the capacity of the tissue to hydrolyze L-LPNA.

Short-chain alkyl esters of L-dopa as prodrugs for rectal absorption.

The bioavailability of L-dopa following rectal administration of a series of short-chain alkyl esters of L-dopa was determined in rats and dogs. The esters were stable (greater than 360 min) to hydrolysis in physiological buffer. In vitro enzymatic hydrolysis of the esters in plasma was species dependent, with the hydrolytic rate being faster in rat plasma (t 1/2 less than 5 min) than dog plasma (t 1/2 = 68-181 min) or human plasma (t 1/2 = 96-238 min). In vivo hydrolysis in dogs, as indicated by the L-dopa plasma profile following intravenous administration of the esters, was very rapid (high extravascular esterase activity). Significant L-dopa bioavailability was observed in rats following rectal administration of the methyl (46%), ethyl (14%), isopropyl (48%), butyl (100%), and 4-hydroxybutyl (13%) esters of L-dopa (rectal L-dopa absorption, less than 5%). In dogs, significant L-dopa bioavailability was also observed for the methyl (28%), isopropyl (30%), butyl (32%), and 4-hydroxybutyl (34%) esters of L-dopa in the presence of carbidopa. The data indicate that these highly water-soluble (greater than 600 mg/ml) esters of L-dopa are potential candidates for controlled-release rectal delivery systems designed to provide more constant plasma L-dopa levels.

Buccal drug absorption. II: In vitro diffusion across the hamster cheek pouch.

The in vitro diffusion of a series of substituted acetanilides across the hamster cheek pouch was studied. The keratinized epithelial layer of the cheek pouch appeared to provide the major barrier to diffusion of these compounds. Linear relationships were found for plots of log epithelial permeability (Pe) versus the log of the octanol-buffer partition coefficient (PCoct; r = 0.712), and log Pe versus the log of the isooctane-buffer partition coefficient (PCiso; r = 0.869). Comparison of the in vitro data with in vivo data obtained in humans using the buccal absorption test showed good correlation (r = 0.878) between Pe and percent buccal absorption in 5 min. These results suggest that diffusion across the hamster cheek pouch may have some utility in the prediction of human buccal absorption.

Application site dependent ocular absorption of timolol.

Ocular absorption of timolol in rabbits was studied after topical ocular administration of 3H-timolol in an eyedrop or in silicone cylindrical devices that released timolol at 7.2 micrograms/h. The devices were applied in either the inferior or superior conjunctival sac. Timolol concentrations were nearly equal in the inferior and superior portions of ocular tissues when the drug was administered in an eyedrop. Administration in the devices resulted in unequal timolol distribution in the cornea, conjunctiva, sclera, and iris-ciliary body. Timolol concentrations were higher in the part of each tissue that was closer to the site of the device application. Unequal concentrations of timolol in the superior and inferior part of the eye and very low timolol concentrations in the aqueous humor indicated that timolol was absorbed mainly via a noncorneal route from the device placed in the inferior conjunctival sac. Induced blinking at one minute intervals did not change ocular absorption of timolol. Compared with inferior conjunctival sac applications, placement of the devices in the superior conjunctival sac resulted in increased corneal and total ocular absorption of timolol as indicated by higher timolol concentrations in the aqueous humor and by a smaller difference between concentrations in the superior and inferior portions of the examined tissues. The application site dependent ocular absorption indicated that controlled release of timolol in the tear fluid did not result in a uniform timolol distribution in the preocular tear fluid of rabbit eyes.

Design and synthesis of a theophylline bonded-phase column for HPLC--application to separation of aromatic carboxylic acids.

A stationary phase has been designed and synthesized in which theophylline residues are covalently bonded to a silica support through an eight carbon hydrocarbon linkage. The phase offers improved resolution in the separation of aromatic carboxylic acids over that available with conventional reversed phase supports. The column is relatively stable. Retention can be modified by adjusting mobile phase composition with respect to pH, electrolyte type and concentration, and organic modifier as well as by manipulating the temperature at which chromatography is carried out. The capacity factors, k', for a series of ring substituted benzoic acids were correlated with the complexation constants previously reported for these compounds with theophylline in bulk phase solution.

Stability of propranolol hydrochloride suspension compounded from tablets.

The stability of a propranolol hydrochloride suspension compounded from commercially available tablets was studied. Propranolol hydrochloride 10-mg tablets were triturated to a powder and incorporated into a commercially available suspension vehicle to yield a suspension with a theoretical propranolol hydrochloride concentration of 1 mg/mL. The suspension was divided into portions and stored in amber glass bottles at either room (25 degrees C) or refrigerated (2 degrees C) temperature for four months. The concentration of propranolol hydrochloride in the samples was determined by a stability-indicating, high-performance liquid chromatographic (HPLC) assay at 0, 30, 60, 90, and 120 days. Also at these times, the pH of the samples was measured, and the samples were inspected visually for evidence of microbial growth and ease of resuspension. The concentration of propranolol hydrochloride in the samples remained within 90% of the initial concentration throughout the study period. No important changes in the pH of the samples and no visible evidence of microbial growth were noted. This extemporaneous suspension of propranolol hydrochloride compounded from commercially available tablets is stable for at least four months when stored at room or refrigerated temperature.

Aspects of the stability of isoindoles derived from the reaction of o-phthalaldehyde-ethanethiol with primary amino compounds.

The stability of a series of fluorescent isoindoles formed under analytical conditions following the reaction of o-phthalaldehyde (OPA) and ethanethiol (ET) with a series of primary amines is reported. Increasing the bulk and degree of substitution of the isoindole N-substituent resulted in substantial increases in isoindole stability. The effects of excess reagents on isoindole stability is examined and OPA is observed to accelerate isoindole degradation whilst ET provides a stabilizing effect. Comparison with previously reported data involving the use of 2-mercaptoethanol revealed that ET clearly forms the more stable isoindole derivatives, i.e. a minimum of five-fold improved stability based on the time for 10% degradation to occur. Identification of the major degradation product together with kinetic data suggests that degradation proceeds via autoxidation.

Rational design and evaluation of improved o-phthalaldehyde-like fluorogenic reagents.

Evidence was presented suggesting that the fluorescent isoindole produced by reaction of o-phthalaldehyde (OPA), ethanethiol, and primary amine was formed by initial imine formation followed by conversion to an alpha-alkylaminobenzylsulfide and subsequent ring closure to form the isoindole nucleus. This mechanism suggested that the minimum structural requirement for condensation to an isoindole was an o-diacyl benzene in which one of the carbonyl groups was aldehydic. A major drawback of OPA as an analytical reagent is the limited stability of the fluorescent 1,2-disubstituted isoindole. Since isoindole instability is related to autoxidation at C-3, the use of o-(formyl) arylketones as alternatives to OPA is attractive in increasing the lifetime of the fluorescent species in that such reagents would form 1,2,3-trisubstituted isoindoles. Two compounds, o-acetylbenzaldehyde (OAB) and o-benzoylbenzaldehyde (OBB), were synthesized and evaluated as potential fluorogenic reagents. Both formed fluorescent products. The rate of formation of isoindole from the latter was too slow to make it of practical analytical value; however, OAB formed isoindoles with t1/2 less than 10 s and offered markedly improved stability over that observed with OPA.

Development of an intravenous formulation for the unstable investigational cytotoxic nucleosides 5-azacytosine arabinoside (NSC 281272) and 5-azacytidine (NSC 102816).

In aqueous solutions 5-azacytosine arabinoside (aza-A) (NSC 281272) exhibits complex and rapid degradation of a type analogous to 5-azacytidine (aza-C) (NSC 102816). Consequently, it is not amenable for use as slow i.v. infusions. This study has determined that both compounds are relatively stable in dry dimethylsulfoxide (DMSO) or dimethylacetamide (DMA). In mixed aqueous-organic solvents, as the water content is reduced the rate of degradation is decreased. Based on these findings, aza-A may be dissolved in DMSO at 100 mg ml-1, sterile filtered, and sealed in ampoules. The contents appear to be adequately stable at 4 degrees C, and may at the time of use be diluted with water to yield a 70% DMSO solution which retains greater than 90% potency for 24 h at 25 degrees C and is compatible with commercially available i.v. infusion tubing. The diluted solution may be added in-line to a flowing i.v. vehicle, resulting in a physiologically acceptable solution in which the drug is unstable (t90 2 h). Its short residence time before reaching the bloodstream precludes any significant loss.

Factors affecting the stability of fluorescent isoindoles derived from reaction of o-phthalaldehyde and hydroxyalkylthiols with primary amines.

The stability of a series of fluorescent isoindole derivatives formed in situ under analytical conditions following the reaction of o-phthalaldehyde (OPA) and 2-mercaptoethanol (2-ME) with a series of primary amines are reported. Increasing the bulk and degree of substitution at C-10 of the resulting isoindole resulted in substantial increases in product stability. The effects of excess OPA and 2-ME on isoindole stability were examined and OPA was observed to catalyze isoindole degradation while 2-ME had no effect. Previously proposed degradation mechanisms were reexamined in light of the present data and an alternate degradation pathway is proposed. 3-Mercapto-1-propanol (3-MP) was found to be a superior thiol for use in the fluorogenic OPA reaction. The OPA/3-MP reagent combination was utilized to derive several amino acids and offered detection limits (S/N = 2) of less than 200 fmol.

Analysis of riboxamide in plasma by high-performance liquid chromatography using automated column switching.

A sensitive and highly specific assay for riboxamide (TCAR) in human and canine plasma is described. The specificity of the procedure is derived from the method of sample preparation and a high-performance liquid chromatographic separation which utilizes the different selectivities of two columns. Partial separation of TCAR from plasma is achieved on a solvent-generated anion exchanger with silica gel as the solid support. The separation is completed by switching the eluent fraction containing TCAR from the first column to a second solvent-generated anion exchanger which has ODS-silica as its support. The relationship between the amount of drug injected and its peak height was linear over wide ranges of concentrations (0-10 micrograms/ml) and injection volumes (20-200 microliter). The limit of detection for TCAR in plasma was 40 ng/ml which can be detected by injecting 200 microliter of processed plasma. The recoveries from plasma were 100.2 +/- 0.9% and 101.3 +/- 2.3% when spiked at the 10 and 1 microgram/ml levels, respectively. The applicability of the method to pharmacokinetic studies was demonstrated by following the plasma levels of TCAR after intravenous administration in the dog.

Tumor concentration of platinum in patients with head and neck cancer.

Simultaneous concentrations of total plasma platinum, filterable platinum, intact cisplatin, and total tumor platinum were measured for 24 hours after an intravenous bolus of cisplatin in five patients. Tumor concentrations of drug were greater than could be explained on the basis of circulating plasma platinum at the time of biopsy. Intratumor platinum concentrations derived from this study provide guidance for selection of the appropriate drug concentrations for in vitro chemosensitivity testing of head and neck cancer in humans.

Evaluation of reductive amperometric detection in the liquid chromatographic determination of antineoplastic platinum complexes.

The usefulness of reductive electrochemical detection at mercury drop electrodes has been determined for platinum complexes separated by solvent-generated anion-exchange high-performance liquid chromatography. Both current-sampled dropping mercury and hanging mercury drop electrodes (DME and HMDE) provide significant advantages over UV absorbance and off-line non-flame atomic absorption detection. The effects of chromatographic and polarographic parameters on analytical system performance have been investigated. By raising the detector cell temperature, the detector response to cis-dichlorodiammineplatinum(II) (DDP) can be shifted anodically to 0.0 V vs. Ag/AgCl, thereby increasing detector selectivity for this compound. The noise-limited minimum detectable quantities of DDP with DME and HMDE are 1.8 ng and 70 pg injected, respectively. DDP can be determined in untreated urine at levels below 100 ng/ml.

High-performance liquid chromatography of cisplatin.

The retention behavior of cisplatin on a variety of stationary phases has been investigated using aqueous mobile phases modified by the addition of various electrolytes and methanol. Cisplatin is poorly retained on reverse-phase or silica columns but satisfactorily retained on chemically bonded or solvent-generated anion exchangers. The retention of the neutral complex on positively charged stationary phases is explained in terms of ion-dipole interactions and rationalized by the application of solvophobic theory. The use of solvent-generated anion exchangers for the analysis of cisplatin offers significant advantages over the chemically bonded system in terms of peak shape, column efficiency, and stability. By the use of column switching and off-line atomic absorption, solvent-generated anion exchange high-performance liquid chromatography (HPLC) is applicable to the determination of cisplatin in urine.

Monitoring the reactions of cisplatin with nucleotides and methionine by reversed-phase high-performance liquid chromatography using cationic and anionic pairing ions.

Methodology, based on reversed-phase high-performance liquid chromatography, is described for monitoring the reactions of cisplatin with DNA, nucleotides, and methionine. Cisplatin was determined in DNA ultrafiltrates on solvent-generated anion exchangers which were prepared by coating the surface of a reversed-phase column with hexadecyltrimethylammonium bromide. These systems were also applicable to studies on the reactions of cisplatin with nucleotides. The retention of the nucleotides studied (5'-AMP, 5'-GMP, 5'-CMP, and 5'-TMP) was described by means of an ion-exchange model and was manipulated by controlling the phosphate concentration in the mobile phase and its pH. The results indicate that cisplatin interacts predominantly with adenosine and guanosine groups on the DNA molecule and that binding is limited by the rate of conversion to an aquated intermediate. Whereas reversed-phase HPLC systems employing cationic pairing ions were applicable to the analysis of mixtures containing cisplatin and anionic solutes, systems employing alkyl sulfonates were required to monitor the reaction of cisplatin with methionine which produces cationic products. Retention, in this latter system, was optimized by the addition of acetonitrile to the mobile phase and by controlling the concentration and chain length of alkylsulfonate in the mobile phase. Although an octadecylsilylsilica, reversed-phase column was preferred for the analytical separation of the methionine-platinum complexes, a polystyrene-divinylbenzene colume was preferred for preparative work.

Cation exchange contribution to the retention of specific quaternary ammonium compounds in reversed-phase high-performance liquid chromatography.

The influence of electrolytes on the retention of organic cationic solutes in reversed-phase high-performance liquid chromatography (RP-HPLC) has been investigated. The effects of the nature and concentration of electrolytes and mobile phase pH on the retention of two model quaternary ammonium compounds were studied on mu-Bondapak C18 stationary phase with aqueous methanolic eluents. The nature and concentration of inorganic cations added to the mobile phase modified the retention of the solutes. The counter anion of the added electrolyte did not perceptibly influence solute retention at constant mobile phase pH, although it did significantly influence solute retention when the electrolytes were added to unbuffered mobile phases. The retention data are consistent with the inclusion of an ion exchange contribution to the retention of cationic solutes in the systems investigated.