The effect of formulations and experimental conditions on in vitro human skin permeation—Data from updated EDETOX database

https://doi.org/10.1016/j.ijpharm.2012.05.012Get rights and content

Abstract

In vitro methods are commonly used in order to estimate the extent of systemic absorption of chemicals through skin. Due to the wide variability of experimental procedures, types of skin and data analytical methods, the resulting permeation measures varies significantly between laboratories and individuals. Inter-laboratory and inter-individual variations with the in vitro measures of skin permeation lead to unreliable extrapolations to in vivo situations. This investigation aimed at a comprehensive assessment of the available data and development of validated models for in vitro skin flux of chemicals under various experimental and vehicle conditions.

Following an exhaustive literature review, the human skin flux data were collated and combined with those from EDETOX database resulting in a dataset of a total of 536 flux reports. Quantitative structure–activity relationship techniques combined with data mining tools were used to develop models incorporating the effects of permeant molecular structure, properties of the vehicle, and the experimental conditions including the membrane thickness, finite/infinite exposure, skin pre-hydration and occlusion.

The work resulted in statistically valid models for estimation of the skin flux from varying experimental conditions, including relevant real-world mixture exposure scenarios. The models indicated that the most prominent factors influencing flux values were the donor concentration, lipophilicity, size and polarity of the penetrant, and the melting and boiling points of the vehicle, with skin occlusion playing significant role in a non-linear way. The models will aid assessment of the utility of dermal absorption data collected under different conditions with broad implications on transdermal delivery research.

Introduction

Skin is in continuous contact with exogenous molecules. Skin's essential role is to protect the body from absorption of exogenous toxic material such as pesticides that target toxicological endpoints and can have local and systemic effects (Nielsen et al., 2004). Therefore, the European Commission program, REACH, requires extensive risk assessments of all existing chemicals, including exposure via dermal contact (Commission of the European Communities, 2003). Skin is also the focus of research by drug formulators as a site of drug administration, both for local dermatologic conditions as well as for systemic delivery due to the advantages it may offer over other routes of drug delivery (Barry, 2007).

A vast number of studies in the past have compared the in vitro and in vivo methods for measuring dermal absorption and have come to the conclusion that properly conducted in vitro measurements can be used to predict in vivo absorption (OECD, 2004a). The OECD Test Guidelines 428 has also confirmed that in vitro studies can predict in vivo absorption when the correct methodology for both tests is used. In vitro methods can vary greatly in terms of the source of skin samples, experimental procedures, and the resulting measurements. These guidelines are flexible in terms of the use of animal or human skin samples (OECD, 2004a, OECD, 2004b). In the literature, many of the reported data pertains to the experiments using artificial skin membranes that mimic human skin. In terms of the human skin samples, the skin can be cadaver human skin or surgically removed skin which may be used fresh as viable skin or after certain period of freezing. These are all sources of variability in the reported results. For example, the absorption of benzoic acid and para-aminobenzoic acid were significantly greater in nonviable, compared with viable, metabolically active hairless guinea pig skin (Nathan et al., 1990). Moreover skin samples can be full thickness or dermatomed with varying thicknesses (Wilkinson et al., 2006).

Apart from the skin sample, the experimental procedures can also influence the results of the in vitro tests. For example, stratum corneum can significantly change its dimensions when exposed for long periods to water (Bouwstra et al., 2003). Experimental approaches vary from studies employing pre-hydrated skin samples, to those using infinite doses which lead to the skin hydration during the period of the experiment, or studies using occlusion of the skin which may lead to some levels of hydration during the experiment, or those employing finite dosing without occlusion which limits the skin hydration. Skin occlusion has been found to enhance the percutaneous absorption of many, but not all topically applied compounds (reviewed by Zhai and Maibach (2001)). On the other hand, unoccluded conditions can simulate the normal exposure situations in everyday life. However, volatile compounds may evaporate under unoccluded conditions and infinite dosing can only take place under occluded conditions (Kligman, 1983, Bronaugh and Stewart, 1985, Baker, 1986). According to OECD (OECD, 2004b), for finite dose experiments, a dose of 1–5 mg/cm2 or 10 μl/cm2 should be spread on the skin surface and for infinite dose experiments, a dose higher than 10 mg/cm2 or 100 μl/cm2 is needed in order to obtain steady state conditions from which the flux and kp can be calculated. In the literature, a full spectrum of application doses can be found with varying duration of exposure and sampling time.

The inter- and intra-laboratory variation in in vitro percutaneous absorption methodology has been investigated to some extent in the past. In a recent study by Van de Sandt et al. (2004) the in vitro absorption of several compounds through human and rat skin were determined in different laboratories. In all laboratories the studies were undertaken according to detailed protocols of dose, exposure time, vehicle, receptor fluid, preparation of membranes and analysis. Results of this study showed noticeable differences that may be attributed to the inter-individual variability in absorption between samples of human skin and differences in skin site and source. Skin thickness only slightly influenced the absorption of benzoic acid and caffeine; however the maximum absorption rate of the most lipophilic compound, testosterone, was clearly higher in laboratories using thin, dermatomed skin membranes.

A vehicle can play a very important role in the penetration of a chemical through the skin. Solubility of a chemical is different in different vehicles hence resulting in different flux and kp values due to varying levels of saturation (Roberts et al., 2002). A vehicle can promote the penetration of a chemical by having low solubility, in this way a chemical will not be retained in the vehicle (Baker, 1986). In case in the vehicle there are components that can interact with the intercellular lipids of the SC then it is possible that permeation may be enhanced or suppressed (Davis et al., 2002). Formulation ingredients can alter the skin penetration of a compound by affecting the barrier properties of the skin or by changing the partitioning of the compound into the SC. The effect of mixture/formulation components on the skin penetration of a compound depends on the nature of the component, i.e. its chemical structure and physicochemical properties. The relationship between chemical structures of the formulation ingredients and the skin penetration modification can be studied quantitatively using quantitative structure–activity relationship (QSAR) techniques (Ghafourian et al., 2004, Ghafourian et al., 2010a, Ghafourian et al., 2010b, Riviere and Brooks, 2005, Riviere and Brooks, 2011).

As an integral part of the human health risk assessment of chemicals and also to be able to aid drug delivery through skin, it is essential to be able to estimate absorption of chemicals via the dermal route. This is because despite the requirement by REACH for extensive risk assessment of chemicals, it is not practical to measure dermal absorption of the many thousands of industrial chemicals. In reality, estimation of skin absorption is complicated due to the inconsistency of the methods and therefore the inconsistent results of in vitro/in vivo tests. The inter-laboratory and inter-individual variations are often high (Van de Sandt et al., 2004, Chilcott et al., 2005) which may be explained by the huge variety of methods and test systems used for skin permeation experiments. Moreover, methods of calculation and interpretation of results from the complex experimental set-up also varies (Henning et al., 2009). The aim of this study was to investigate the effects of experimental conditions such as membrane thickness, occlusion, hydration, vehicle ingredients and mode of exposure (finite or infinite dosing) on the skin permeation flux. This was achieved through the use of statistical techniques employing a large dataset extracted from EDETOX database and collated from more recent publications. The dataset was large enough to investigate statistically the effects of these variable experimental conditions combined with QSAR linking the skin flux to the chemical structures of the penetrants and the physico-chemical properties of the vehicle mixtures.

Section snippets

The dataset

The in vitro flux of chemicals from human skin measured by flow-through or static cells were obtained from the recent literature (2001–2010) and EDETOX (Evaluations and Predictions of Dermal Absorption of Toxic Chemicals) database (EDETOX, 2010). The compiled data is available as Supplementary material I. The EDETOX database contains data from in vitro and in vivo percutaneous penetration studies involving use of different species, cell types and chemicals with a total of 2501 records taken

Results

For the development of predictive models and investigation of the effects of various experimental, vehicle or permeant variables, the collated dataset was first refined and the outliers were removed. The investigation was then focused on the development of a series of models with specific emphasis on the descriptors of in vitro experimental conditions and vehicle properties. Finally the models were validated and compared in terms of the accuracy of the skin flux predictions.

Discussion

Inter-laboratory and inter-individual variations are very common in the in vitro measures of skin permeation. This has been attributed to a number of experimental variables including differences in skin samples’ thickness, skin hydration, occlusion of the skin, infinite or finite dosing and vehicle ingredients. The dataset gathered here provided an excellent resource for development of QSAR models which also incorporate the effect of various experimental variables. The models can elucidate the

Notation

Model parametersDescription
[donor]donor concentration (μg/ml)
BP  MP(mix)Difference between the boiling and melting points of the mixture (donor phase)
chi0Zero order molecular connectivity indexa
chi0vZero order valence molecular connectivity indexa
chi1v_CFirst order carbon valence connectivity indexa
fiABFraction of molecules ionised as anion and cation at pH 7.4
GCUT_PEOE_1The GCUT descriptors are calculated from the eigenvalues of a modified graph distance adjacency matrix. Each ij entry of the

Acknowledgement

Eleftherios Samaras was supported by Medway School of Pharmacy.

References (59)

  • A. Henning et al.

    Potential pitfalls in skin permeation experiments: influence of experimental factors and subsequent data evaluation

    Eur. J. Pharm. Biopharm.

    (2009)
  • M.E. Johnson et al.

    Permeation of steroids through human skin

    J. Pharm. Sci.

    (1995)
  • C.T. Jung et al.

    In vitro and in vivo percutaneous absorption of catechol

    Food Chem. Toxicol.

    (2003)
  • Y.B. Liou et al.

    Construction of a quantitative structure-permeability relationship (QSPR) for the transdermal delivery of NSAIDs

    J. Control. Release

    (2009)
  • L.M. Monene et al.

    Percutaneous absorption of cyclizine and its alkyl analogues

    Eur. J. Pharm. Sci.

    (2005)
  • S. Patil et al.

    Epidermal enzymes as penetration enhancers in transdermal drug delivery

    J. Pharm. Sci.

    (1996)
  • J.E. Riviere et al.

    Predicting skin permeability from complex chemical mixtures

    Toxicol. Appl. Pharmacol.

    (2005)
  • P. Santos et al.

    Oxybutynin permeation in skin: the influence of drug and solvent activity

    Int. J. Pharm.

    (2010)
  • P. Sartorelli et al.

    Percutaneous penetration studies for risk assessment

    Environ. Toxicol. Pharmacol.

    (2000)
  • R.J. Scheuplein et al.

    Percutaneous absorption of steroids

    J. Invest. Dermatol.

    (1969)
  • M. Sznitowska et al.

    Studies on the effect of pH on the lipoid al route of penetration across stratum corneum

    J. Control. Release

    (2001)
  • H. Tang et al.

    Prediction of steady-state skin permeabilities of polar and nonpolar permeants across excised pig skin based on measurements of transient diffusion: characterization of hydration effects on the skin porous pathway

    J. Pharm. Sci.

    (2002)
  • J.J.M. Van de Sandt et al.

    In vitro predictions of skin absorption of caffeine. Testosterone and benzoic acid: a multi-centre comparison study

    Regul. Toxicol. Pharmacol.

    (2004)
  • H. Baker

    The skin as a barrier

  • B.W. Barry

    Transdermal drug delivery

  • T. Bouwman et al.

    Improving the applicability of (Q)SARs for percutaneous penetration in regulatory risk assessment

    Hum. Exp. Toxicol.

    (2008)
  • D. Bucks et al.

    Effects of occlusion

  • Commission of the European Communities. Proposal for a Regulation of the European Parliament and of the Council...
  • ChemBioFinder. CambridgeSoft. Available from: http://www.cambridgesoft.com/databases (accessed November 2010–February...
  • Cited by (36)

    • The importance of metrological tools to implementation of alternative method OECD TG 428

      2022, Toxicology in Vitro
      Citation Excerpt :

      It is a European database where measurement results from different laboratories can be organized by the type and origin of samples, thickness, and receptor fluid composition, among other conditions mentioned above (EDETOX database, 2022; OECD, 2004). In this paper, we focus on the lack of comparability of the results regarding the permeation parameter/flow rate, although it is widely discussed in the literature (Ackermann et al., 2010; Duracher et al., 2009; Jacques-Jamin et al., 2017; Samaras et al., 2012; Van De Sandt et al., 2004). Comparability is an essential requirement of an international system in which measurement results are universally accepted.

    • Mechanistic skin penetration model by the COSMOperm method: Routes of permeation, vehicle effects and skin variations in the healthy and compromised skin

      2019, Computational Toxicology
      Citation Excerpt :

      More precisely, a 1,2-dimyristoyl-sn-glycero-3-phosphocholine (DMPC) membrane, as simulated by Sven Jakobtorweihen [46,47], plus an electrostatic membrane potential [48]. The experimental data points are extracted from the EDETOX database [49] in the updated version [7,17]. The following criteria were applied: a) only in vitro/ex vivo human data are used b) with reported experimental kp values, c) with a reported membrane type stratum corneum or isolated epidermis and d) information on the vehicle provided.

    View all citing articles on Scopus
    View full text