Prodrugs for dermal delivery

Abstract

The recent review by J. Hadgraft (1985, Prodrugs and skin absorption. In: H. Bundgaard (Ed.), Design of Prodrugs, Elsevier Biomedical Press, Amsterdam, pp. 271–289) presents a good overview of the use of prodrugs to enhance skin absorption. However, at that time there was little information about trends in solubilities of, and in abilities to deliver parent drugs by homologous series of prodrugs available for analysis so no correlations between solubilities of the prodrugs and their abilities to deliver their parent drugs were made. Also, although applications of quite a few different types of promoeities were described, there was no attempt to describe reported and potential prodrug strategies for each type of functional group. In the present review an attempt has been to address those latter two points. Thus, strategies for the introduction of promoeities onto primary and secondary amines, amides, imides, hydroxyls, thiols, carbonyls and carboxylic acid functional groups have been described. Also, for several of the strategies, where results from homologous series of prodrugs were available for analysis, two conclusions were reached. First, although an increase in lipid solubility due to transient masking of a polar functional group almost always results in enhanced dermal delivery of the parent drug, in order to optimize delivery, it is necessary to use the members of the series that are more water soluble than the parent drug or that are the more water-soluble member(s) of the series. Second, for dermal delivery from a given vehicle, there is a good correlation between the log experimental permeability coefficients ( P) and the calculated solubility parameters ( δ i) of the drugs and prodrugs. For many of the strategies that have been described, only one or two examples were available to show the effect that implementing a particular prodrug strategy had on solubilities and hence on ability to enhance dermal delivery. However, application of conclusions based on the results from the homologous series of prodrugs to these latter strategies should allow one to optimize dermal delivery for them as well. The bases for the first conclusion can be best understood in terms of two recent publications. The first is a report by D.W. Osborne (1988, Theoretical considerations of transdermal delivery: computational optimization of the drug's physical chemical properties, ACS 195th National Meeting, Toronto, Canada, Abstract, Medi 10) on the effect of water solubility and partition coefficient ( K) on the transdermal flux of a series of hypothetical molecules of constant molecular weight. A group of contour plots of flux for each molecular weight were obtained which showed that, in order for increases in K to result in increased flux, the water solubility had to remain the same and vice versa. However, extrapolation of these results to the situation where a homologous series of prodrugs has been synthesized and the optimum member of the series needs to be identified leads to the same conclusion reached in this review. For example, in the α-acyloxyalkyl series as the acyl chain length is increased water solubility goes down as lipid solubility and hence K goes up. The result is at best no change in the rate of delivery of the parent drug by the first few members of the series of prodrugs is observed, but as the chain length increases further and eventually lipid solubility also starts to decrease along with water solubility the rate of delivery has to decrease. Similary, G.B. Kasting, R.L. Smith and E.R. Cooper (1987, Effect of lipid solubility and molecular size on percutaneous absorption. In: B. Shroot and H. Schaefer (Eds.), Pharmacology and the Skin, S. Karger, Basel, pp. 138–153) have shown that the two parameters that are most important for determining the transdermal delivery of molecules are melting point and size (or molecular weight). The solubility of the permeant molecule in skin lipids is estimated from ideal solution theory where the ideal solubility decreases almost exponentially with increasing melting point. Thus the dependence of delivery on melting point. Also, for a homologous series the shorter chain initial members of the series are the smaller molecules. Hence they should diffuse through the skin faster. These smaller molecules are also the more water-soluble members of the series. Again, the above conclusions support the conclusions of this review.