Prediction of drug distribution into human milk from physicochemical characteristics

Decisions about the safety of breast feeding during maternal ingestion of drugs require knowledge of the amount of drug which might be present in the milk. For many drugs this has not been studied, and mothers are usually advised against breast feeding. In many cases this is undoubtedly unnecessary, as the total dose to which the baby is exposed is often negligible. It would be very helpful, therefore, to be able to predict the approximate amount of drug which might be present in milk. Existing theory of pH partitioning enables estimation of the distribution of unbound drug, i.e. milk: plasma unbound ratios. However, these ratios are poor estimates of the concentration ratios for whole milk, because whole milk contains proteins and lipid in which drugs will distribute in amounts which depend on their particular physicochemical properties. To predict the milk: plasma concentration ratios for whole milk the amount of drug present in the protein and lipid phases must be considered along with the unbound drug distribution. A 'phase distribution model' has therefore been developed which permits estimation of whole milk: plasma concentration ratios. The model requires a knowledge of the unbound drug concentration ratio, the plasma and milk unbound fractions and the milk lipid: ultrafiltrate partition coefficient. Evaluation of the model by comparison of predicted whole milk ratio values with literature milk: plasma area under the curve (AUC) ratios indicated a trend to overprediction for acidic and neutral drugs and underprediction for basic drugs. Transformation of the phase distribution equation by taking logarithms results in a relationship which can be analysed by multiple linear regression to derive predictive equations for acidic and basic drugs which take into account the relative contributions of each component of the model. Regression of the logarithms of the literature milk: plasma AUC values against the independent variables resulted in good correlations for acidic and basic drugs. The independent variables explained 93.1% and 82.9% of the variance in the values for acidic and basic drugs, respectively, with random scatter of residuals. The equations, together with those to predict unbound fractions of drug in milk and milk lipid: ultrafiltrate partition coefficients, enable the ratio of the milk: plasma AUCs to be estimated for any acidic or basic drug for which the distribution into human milk is not known, using the pKa, octanol: water partition coefficient and plasma protein binding values of the drug. The data set for neutral drugs (n = 3) was too small to develop a correlation equation.(ABSTRACT TRUNCATED AT 400 WORDS)     

The influence of drug partition coefficient on follicular penetration: in vitro human skin studies.

The aim of this study was to employ the novel skin sandwich system in order to quantify the influence of the octanol-water partition coefficient on follicular drug absorption in human skin. To this end, seven different drugs - estradiol, corticosterone, hydrocortisone, aldosterone, cimetidine, deoxyadenosine and adenosine - exhibiting a wide range of log octanol-water partition coefficients (logK(o/w)) but relatively similar molecular weights were selected as candidate solutes. Application of the skin sandwich technique yielded an interesting relationship between % follicular contribution and logK(o/w). The follicular contribution to total flux was small (4 and 2%) for the two most lipophilic drugs but varied between 34 and 60% for the remaining drugs of intermediate and low logK(o/w) values. Lipophilicity seems to be an important modulator of drug absorption into follicular orifices only above a critical logK(o/w) threshold. Below this critical logK(o/w) value, lipophilicity does not apparently influence the follicular contribution in an obvious way and the process is probably governed by other molecular properties. Identification of these other active properties would require performing this kind of a study on a much larger set of candidate drugs.     

In vitro percutaneous absorption of prednisolone derivatives based on solubility parameter

A series of ester derivatives of prednisolone (I–VII) with various lipophilicities was synthesized to investigate their in vitro percutaneous absorption and their distribution and accumulation in the skin. Experimental findings were supported with theoretical calculations, using the solubility parameter as an indicator of the lipophilicity of the derivatives. The solubility parameters of the derivatives were well correlated with their partition coefficients in an octanol-water system and increase in lipophilicity was well correlated with the theoretical and experimental values of drug distribution into the skin. Drug distribution into the skin was increased, but the drug diffusion rate was decreased with increasing derivative lipophilicity. These findings indicated that accumulation of the derivatives in the skin increased with increasing derivative lipophilicity. This suggests that retention of prednisolone derivatives in the skin is a function of their solubility parameters and that derivatives are partitioned in the skin relative to their lipophilicities.     

Evidence that drug flux across synthetic membranes is described by normally distributed permeability coefficients.

Over recent decades, the use of in vitro diffusion cell studies to assess skin permeability has evolved into a major research tool, providing key insights into the relationships between skin, drug and formulation. Sometimes, such studies involve synthetic membranes as this approach can yield useful inferences with respect to drug-skin partitioning and diffusion phenomena. Yet despite the popularity of such studies, it is still not at all known whether typical solute transport across synthetic barriers results in a normal distribution of permeability coefficients or alternatively some type of skewed distribution. The present study aims to shed light on this issue. To this end, five compounds (testosterone, oestradiol, corticosterone, aldosterone and adenosine) exhibiting a broad range of octanol-water partition coefficient values were selected as test penetrants. The protocol involved taking multiple replicate measurements of each drug's passive steady state flux through poly(dimethylsiloxane) membrane. Each penetrant's resultant permeability coefficient database was subjected to a Kolmogorov-Smirnov (KS) test for normality. It was found that the permeability coefficients of all five drugs were distributed in a Gaussian-normal fashion. The theoretical significance and practical impact of these findings are discussed.     

Estimation of the effect of NaCl on the solubility of organic compounds in aqueous solutions

The Setschenow constant, K(salt), of a nonelectrolyte in a NaCl solution is shown to be related to the logarithm of its octanol-water partition coefficient, log K(ow), determined by K(salt) = A log K(ow) + B, where K(ow) is the octanol-water partition coefficient of the solute and the coefficients A and B are constants. The values of A and B were empirically determined from literature data for 62 organic compounds and validated for a test set of 15 compounds including several drugs.     

The binding of drugs to different polar lipids in vitro

For evaluation of the physico-chemical basis underlying the drug-induced generalized lipid storage disease, the equilibrium distribution of radioactively labelled amphiphilic drugs between a water phase and liposomes was determined. Liposomes were prepared from sphingomyelin (SM), phosphatidylcholine (PC), phosphatidylethanolamine (PE), and phosphatidylserine (PS); the extent of binding of the drugs occurred in increasing order SM ～ PC < PE < PS. A strong correlation was found between the octanol-water partition coefficient and the PC-water coefficient for monovalent cationic drugs; the absolute values of both partition coefficients resembled each other very closely, suggesting that the hydrophobic forces are mainly responsible for the binding of amphiphilic drugs to PC. The higher extent of binding to PS does not result from higher afiinities of the drugs to PS but rather from the higher capacity of PS-liposomes as compared with that of PC- or SM-liposomes. The divalent cationic drug chloroquine displayed particularly high binding to the anionic lipids PS and to gangliosides as compared with the monovalent drugs. This observation might help to explain the ultrastructural and biochemical findings that chloroquine induces a remarkable storage of anionic lipids upon chronic treatment of animals.     

Influence of lipophilicity and lysosomal accumulation on tissue distribution kinetics of basic drugs: a physiologically based pharmacokinetic model

This paper examines the role of lipophilicity in the tissue distribution kinetics of basic drugs. Basic drugs have a large distribution volume and are distributed widely in various tissues in the following order: lung, fat, heart, kidney, brain, gut, muscle and bone. The fat volume in the whole body influences the disposition kinetics. There is a good correlation in various tissues between the tissue-plasma concentration ratio and the octanol-water partition coefficient among various drugs. We constructed a physiologically-based pharmacokinetic model on the basis of drug lipophilicity and found that drug distribution decreased when NH4Cl was administered concomitantly. In regards to subcellular distribution, the relative specific contents of chlorpromazine, imipramine and biperiden with respect to the protein in lysosomes were 7.3, 9.6 and 4.2, respectively, while those in other subcellular organella, including mitochondria, were only 0.4-1.7, indicating preferential accumulation of these drugs in lysosomes. The uptake of basic drugs into lysosomes depended on both intralysosomal pH and drug lipophilicity. As the lipophilicity of the basic drugs increased, they accumulated more than would have been predicted from the pH-partition theory and raised the intralysosomal pH more potently, probably owing to their binding with lysosomal membranes, with or without intralysosomal aggregation. We conclude that the distribution kinetics of basic drugs is driven by drug lipophilicity and uptake into lysosomes, and these phenomena provide a possible basis for drug interaction in clinical treatments.     

Rapid determination of liposome-water partition coefficients (Klw) using liposome electrokinetic chromatography (LEKC)

Liposome Electrokinetic Chromatography (LEKC) provides a simple and facile approach for determining liposome-water partition coefficients, Klw. LEKC is a Capillary Electrophoresis (CE) technique where liposomes are incorporated into a buffer solution and act as a pseudostationary phase providing sites of interaction for solutes. The retention factors of solutes in LEKC are directly proportional to Klw. This article describes how LEKC can be used to determine Klw for both neutral and charged solutes. The Klw values for a group of neutral aromatic compounds, beta-blockers, and other drugs are reported. In addition, the usefulness of two quantitative structure partition relationships (QSPR) for estimation of Klw is demonstrated. One is the logarithmic linear relationship between liposome water and octanol-water partition coefficients (Pow). The other is the Linear Solvation Energy Relationship (LSER). The calculated Klw values from the two QSPR agree nicely with the observed values and with one another.     

Narcotic and narcotic antagonist pKa''s and partition coefficients and their significance in clinical practice

The pKa's, partition coefficients and drug distribution coefficients (apparent partition coefficients) have been investigated for a number of narcotics and where possible for their congener narcotic antagonist. These studies were carried out by a novel microelectrometric titration technique as a function of temperature and pH. This method enables one to determine not only the dissociation constants to deconvolute overlapping pKa's, but also to determine the solubilities and oil-water distribution of these various drugs. The drug distribution coefficients displayed marked sensitivity to pH at values which span the range of attainable human physiological pH values. This has significant pharmacological implications for proper choice and scaling of drug dosages under various clinical situations among which are cited hyperventilation under a general anesthetic while concomitantly under a narcotic analgesic, obstetetrical analgesia, and medical and anti-abuse usage of narcotic antagonists. The partition coefficients and drug distribution coefficients were noticeably different at 20degreesC (where such measurements are customarily made) from those at 37degreesC (body temperature). Furthermore, various drugs exhibit very non-equivalent increases in drug distribution coefficients with increasing temperature, ranging from 21% for naltrexone. This non-regularity indicates that it will not be valid to extrapolate by any constant factor the measurements made at lower temperatures. Even the true partition coefficients increase with temperature from 20degrees to 37degreesC.     

Microelectrometric titration measurement of the pKa's and partition and drug distribution coefficients of narcotics and narcotic antagonists and their pH and temperature dependence.

The pKa's, partition coefficients, and drug distribution coefficients (apparent partition coefficients) have been investigated for a number of narcotics and, where possible, for their congener narcotic antagonists. These studies were carried out by a microelectrometric titration technique as a function of temperature and pH. This method enables one to determine not only the dissociation constants to deconvolute overlapping pKa's, but also to determine the solubilities of oil-water distribution of these various drugs. The drug distribution coefficients displayed marked sensitivity to pH at values which span the range of attainable human physiological pH values. This has significant pharmacological implications for proper choice and scaling of drug dosages under various clinical situations. The partition coefficients and drug distribution coefficients were noticeably different at 20 degrees (where such measurements are customarily made) than at 37 degrees (body temperature). Furthermore, various drugs exhibit very nonequivalent increases in drug distribution coefficients with increasing temperature, ranginf from 21% for morephine to 200% for naltrexone. This nonregularity indicates that it will not be valid to extrapolate by any constant factor the measurements made at lower temperatures. Even the true partition coefficients increase with temperature from 20 degrees to 37 degrees. There is more of a difference in the drug distribution coefficients for naloxone and naltrexone than might have been expected from the similarities in their structures with naltrexone being significantly less lipophilic than naloxone. This would imply that this would lead to naloxone having a more rapid onset for antagonist activity and likewise a shorter duration of action than naltrexone.     

Corneal penetration behavior of beta-blocking agents I: Physiochemical factors

Rabbit corneas were excised and mounted in a chamber to determine the permeability characteristics of a group of beta-blocking agents which varied in octanol-water partitioning over a fourfold logarithmic range. From the permeability rate at steady state, permeability coefficients (pH 7.65) were determined. For each drug the distribution coefficient and pKa were measured, permitting the partition coefficients to be estimated. Various correlations were determined for the log permeability coefficient as a sum of log functions of the partition (or distribution) coefficient, molecular weight, and/or degree of ionization. The best fit, as judged by a high correlation coefficient (r = 0.9756) and lack of systematic deviation, was represented by: log PT = 0.623 log DC - 0.108(log DC)2 - 5.0268.     

Activated carbon as a biological model: comparison between activated carbon adsorption and oil-water partition coefficient for drug activity correlation

Octanol-water partition coefficients have been used as a principal parameter for the analysis of quantitative structure-activity relationships (QSAR). With ionized drugs, however, the correlation is not impressive. Although clinical local anesthetics exist mainly in the ionized form at physiological pH, the nerve-blocking potency is correlated with the oil-water partition coefficients of the un-ionized species. In the present report, the adsorption of 18 local anesthetics onto activated carbon surfaces was compared with oil-water partition coefficients for correlation with the minimum nerve-blocking (MBC) potency. The log octanol-water partition coefficients showed large differences between the charged and uncharged species (lidocaine: 0.49 and 1.82, procaine: -0.56 and 1.74, respectively), whereas the log carbon surface adsorbabilities showed much smaller differences (lidocaine: 3.00 and 3.04, and procaine 2.95 and 3.00, respectively). An excellent correlation was found between the log carbon surface adsorption and the log MBC values in both ionized and un-ionized drugs.     

In‐vitro permeation of drugs into porcine hair follicles: is it quantitatively equivalent to permeation into human hair follicles?

It is already well-established that the general permeability properties of porcine skin are close to those of human skin. However, very little is known with respect to drug absorption into hair follicles and the similarities if any between the two types of tissue. The aim of this study was to use the skin sandwich system to quantify follicular drug absorption into porcine hair follicles. To our knowledge, this is the first time that the skin sandwich has been extended to porcine tissue. For this purpose, seven different drugs -- estradiol, corticosterone, hydrocortisone, aldosterone, cimetidine, deoxyadenosine and adenosine -- exhibiting a wide range of log octanol-water partition coefficients (log K(o/w)), but comparable molecular weights, were chosen as candidate solutes. The results showed a parabolic profile with maximal follicular contribution occurring at intermediate log K(o/w) values. Linear regression analysis indicated that the follicular contributions in porcine skin correlated well with previously published follicular contributions in human skin (r(2) = 0.87). The novelty of this research is that we show that porcine tissue is a good surrogate for modelling human skin permeability within the specific context of quantifying drug absorption into hair follicles.     

Relationships between the Hepatic Intrinsic Clearance or Blood Cell-Plasma Partition Coefficient in the Rabbit and the Lipophilicity of Basic Drugs

The relationships between drug lipophilicity and hepatic intrinsic clearance (CL_int,h) or red blood cell-plasma partition coefficients (D) have been elucidated for ten highly lipophilic basic drugs with apparent octanol-water partition coefficients at pH 7.4 (P_{app, oct}) of 150 or above. The true octanol-water partition coefficients of the non-ionized drugs (P_oct) were used to determine CL_int,h and D for the unbound drugs (CL_int,h,f and D_f, respectively), and CL_int,h,f and D_f for the non-ionized and unbound drugs (CL_int,h,fu and D_fu, respectively). The total clearance values were determined at steady state by infusion studies of individual drugs in rabbits. There was better correlation between log P_oct, and log CL_int,h,fu (r = 0.974) than between log P_oct, and log CL_int,h,f (r = 0.864). The D values were calculated from the blood-plasma concentration ratio. There was a better correlation between log P_oct and log D_fu (r = 0.944) than between log P_oct, and log D_f (r = 0.612). The regression equations obtained were CL_int,h,fu = 0.0875 × P_oct^1.338 and D_fu = 0.0108 × P_oct^0.970, respectively. These results show that the CL_int,h and D of highly lipophilic basic drugs can be predicted from P_oct by taking f_u into consideration. By applying these parameters to a physiologically based pharmacokinetic model it might be possible to predict the pharmacokinetics of unknown basic drugs.     

A simple, predictive, structure-based skin permeability model

By an extension of our simple, molecular size-based model recently developed to describe octanol-water partition coefficients, we were able to obtain an entirely structure-based model that seems well suited to describe human skin permeability data. The corresponding equations not only eliminate the physicochemical interrelatedness of the parameters of the original Potts & Guy approach that was obtained from similar considerations, but also maintain its elegant simplicity and are consistent with a basic physicochemical model of the related phenomena. As the new model is structure based and fully computerized, it allows direct estimation of skin permeability for any molecule of known structure without the need to obtain octanol-water partition coefficients or other experimental data.     

Prediction of drug loss from PVC infusion bags.

PVC:water partition coefficients for a series of 13 drugs have been calculated from literature data and a high degree of correlation with octanol:water partition coefficients demonstrated. The resulting model, log PPVC = -0.35 + 0.69 log P. (r2 = 0.88) has been prospectively tested with 10 drugs. All the test drugs were within the 95% confidence intervals associated with predicted log PPVC values consistent with a valid model. In practice, predicted log PPVC values may be used to estimate drug loss from the aqueous phase of PVC bags at equilibrium. Equations are described which enable calculation of likely drug loss from 100, 500 and 1000 mL PVC bags. It is recommended that this approach is used to identify drugs which are unlikely to be significantly absorbed into PVC.     

Effect of 1-alkyl- or 1-alkenylazacycloalkanone derivatives on the penetration of drugs with different lipophilicities through guinea pig skin

The percutaneous penetration-enhancing effects of 1-dodecyl- (azone), l-geranyl-, and 1-farnesylazacycloheptan-2-one were investigated using seven penetrants having a wide range of n-octanol-water partition coefficients. The penetration of the drugs from water vehicle (aqueous system) and ethanol vehicle (ethanolic system) through excised guinea pig skin was increased by pretreatment with the enhancers. Large enhancement was observed for the drugs, such as 5-fluorouracil and 6-mercaptopurine, with n-octanol-water partition coefficients of approximately unity. The penetration profiles were analyzed based on a one-layer skin model. Two parameters corresponding to the drug diffusivity and partitioning into the skin were obtained. In the aqueous system, the partitioning of drugs into the skin was increased by pretreatment with the enhancers. This led to an increase in drug penetration and accumulation in the skin; diffusivities were little affected. From these parameters, the drug amounts in the vehicle and the skin were well estimated for drugs having partition coefficients of less than 1. In the ethanolic system, the enhancement was far less than that observed in the aqueous system.     

Comparison of calculated versus measured partition coefficients of some phenyl beta-D-glucopyranosides

Experimentally determined octanol-water partition coefficient values of substituted phenyl beta-D-glucopyranosides are compared with the calculated values using the computer program CLOGP. The systematic deviation of the calculated values from the measured ones in this series suggests that caution is required when calculations are performed on classes of compounds where many of the partition coefficients have not been experimentally determined.     

Drug distribution within human milk phases

Phase distribution and protein binding of drugs in human milk have been measured. The analytical method is reproducible, rapid, and requires only small sample volumes. Five drugs were studied: diazepam, phenobarbital, warfarin, phenytoin, and disopyramide. Experiments were carried out at 37 degrees C on milk samples with variable fat and protein contents. Results for the distribution of drugs between the skimmed-milk phase and fat-rich phase are presented, as well as the results of the dialysis of drugs in skimmed milk. It is shown that, among the physicochemical properties of a drug, the lipid solubility seems to be the most important property for predicting variations in drug concentrations in milk. The potential significance of the findings with respect to in vivo distribution of drugs into human milk is discussed.