Document Type


Date of Degree

Fall 2015

Degree Name

PhD (Doctor of Philosophy)

Degree In

Pharmaceutical Sciences and Experimental Therapeutics

First Advisor

Douglas R. Flanagan


Solid polymer-drug dispersions are used to prepare and stabilize amorphous forms of poorly soluble drugs as a means of improving drug solubility, dissolution and bioavailability. Despite many reports on this subject, solid dispersion dissolution mechanisms have not been well understood. An early study was reported by Simonelli, Mehta and Higuchi (SMH) in 1969 and has served as a model for dispersion dissolution behavior. These authors proposed a dissolution model (SMH) which gave good agreement between their experimental results and model predictions for one drug and one type of PVP.

Few researchers have applied this traditional approach (SMH) in a systematic fashion to solid dispersion systems. One difficulty is obtaining parameters needed for predictions such as polymer diffusion coefficient, diffusion layer thickness or other pertinent parameters. In this work, a general model has been developed based on the concepts in the traditional approach (SMH) and simulations with this model were performed to show how dispersion dissolution rates change with system variables. Such simulations showed underestimation of dissolution rates resulted when compounds had low solubility.

In this work, solid dispersion dissolution behavior was studied systematically with a homologous compound series (alkyl-p-aminobenzoate esters, or PABA esters) and three polyvinylpyrrolidone (PVP) molecular weights (K15, K30 and K90). The PABA esters with varying solubility used in this study were methyl PABA (MePABA), ethyl PABA (EtPABA), propyl PABA (PrPABA) and butyl PABA (BuPABA). Six solid dispersions for each PABA ester and PVP (weight ratios of PVP:PABA ester 20:1, 10:1, 6:1, 3:1, 4:1 and 2:1) were prepared by a solvent evaporation method. Solid dispersions were obtained and their amorphous character confirmed by powder X-ray diffraction (PXRD) and differential scanning calorimetry (DSC). Intrinsic dissolution rates for these dispersions were obtained in water with a rotating disc dissolution system. Both dissolution rate of drug (PABA ester) and carrier (PVP) were measured to obtain more information on which to evaluate the release behavior. Measuring the dissolution of the polymer (dispersion agent) and drug is unique in this work and has not been done in most other reported studies.

For the more soluble PABA esters (i.e., MePABA, EtPABA and PrPABA), as drug loading increased, PABA ester dissolution rates first increased and then decreased to that of the pure drug for PVP K15 and K30 dispersions. For K90 systems, drug dissolution rates were below pure drug rates and increased steadily as drug loading increased, eventually reaching that of the pure drug. On the other hand, PVP dissolution rates decreased constantly as drug content increased for all three PVP grades. However, the decrease in polymer dissolution was more pronounced for the lower molecular weight PVPs (K15 and K30) than the higher molecular weight PVP (K90). Comparison of drug and polymer dissolution behavior indicated that congruent release of both components occurred when drug loading was low (< 15%). As drug loading increased, more deviation from congruent release behavior was observed. For BuPABA, the least soluble PABA ester, precipitated BuPABA solid accumulated on the disc surface during dissolution.

PABA ester relative dissolution rates were calculated and compared with the predictions from the developed general model (based on assumptions in the traditional approach). Such predictions correlated well with experimental results at high drug loadings (i.e., >25%) but at low drug loadings (i.e., <25%) there was inconsistent correlation between experimental and predicted results. A new model was developed in which dispersion systems were generally classified into two regions: a carrier-controlled region and a drug-controlled region. Congruent release was predominated in the carrier-control region and pure drug release occurred in the drug-controlled region. The results showed the new model offered better agreement with experimental results.

Public Abstract

Solid drug-polymer dispersions are used to prepare and stabilize amorphous forms of poorly soluble compounds. These dispersions have attracted considerable interest as a means of improving drug solubility, dissolution and bioavailability. Despite many reports on solid dispersions, the drug release mechanism from dispersion systems has not been well understood.

In this study, a systemic approach was employed to better elucidate solid dispersion dissolution mechanism(s). A homologous compound series (alkyl-p-aminobenzoate esters, or PABA esters) was used together with three different molecular weights of polyvinylpyrrolidone (PVP K15, K30 and K90). Intrinsic dissolution studies were performed on the PABA ester-PVP dispersions. Both drug and carrier dissolution were followed simultaneously which is a unique aspect of this study; in most other such studies, only drug dissolution was monitored.

The results showed that the PABA ester-PVP solid dispersion intrinsic dissolution could be generally categorized into two regions: a polymer-controlled region and a drug-controlled region, which depends on drug loading, drug solubility or PVP used in the dispersion. In the polymer- controlled region, drug dissolution was mainly influenced by drug loading as well as polymer dissolution which depended on the PVP grade used. In the drug-controlled region, drug dissolution was largely independent of carrier and was equivalent to that of the pure drug. The model developed in this study described the dissolution behavior of these dispersions fairly well.




xxiii, 161 pages


Includes bibliographical references (pages 157-161).


Copyright 2015 Yi Wu