Use this URL to cite or link to this record in EThOS:
Title: In vitro and in silico profiling to assist pharmaceutical development
Author: Perrier, Jean-Baptiste Jeremy
ISNI:       0000 0004 7972 2793
Awarding Body: University of Strathclyde
Current Institution: University of Strathclyde
Date of Award: 2018
Availability of Full Text:
Access from EThOS:
Access from Institution:
During the pharmaceutical development process of a new drug it is necessary to fully elucidate the physico-chemical properties of the molecule in order to find the appropriate formulation. For drugs intended to be administered orally water-solubility is a an essential requirement. The development of high-throughput screening and combinatorial chemistry gave a rapid access to a lot of potential drug candidates but as a consequence led to new challenges. On one hand there is an increasing number of molecules with poor water solubility which can be addressed with adequate in vitro testing. On the other hand studies demonstrated that poor pharmacokinetic behaviour was responsible for the majority of drug candidate failures. Therefore the challenge is to be able to keep up with the rate of production to study the in vivo behaviour of so many drugs more rapidly and efficiently. The emergence of physiologically based pharmacokinetic modelling provides a potential solution by making use of the physico-chemical information along with the human physiology to simulate the clinical performance. In this work an experimental approach is presented to study the in vitro solubility of poorly soluble drugs. A new biorelevant medium simulating the fasted and fed intestinal states was developed to examine the feasibility of merging the individual fasted and fed studies into one single experiment. The purpose was to forecast the parameters influencing solubility and anticipate the solubility behaviour such as potential significant food effects. Results showed that acidic drugs were greatly driven by pH and oleate while for neutral and basic drugs a combination of three factors (pH, bile salt and sodium oleate) and their interactions were dominant. Solubility testing in biorelevant media has been accepted as a reliable predictive tool of in vivo solubility. The prediction of better estimates of in vivo behaviour allowed BCS class II drugs to be reconsidered provided that an adequate formulation is designed. This is applied in the recent developability classification system (DCS). This classification incorporates an estimate of the fasted intestinal solubility along with the dose/solubility ratio of the drug and allows a better understanding of the factors limiting oral absorption. This classification adds a distinction between dissolution-rate limited drugs and solubility-limited drugs which provides a very useful tool for formulators in the early stage of development. This approach is presented in this work on a set of BCS class II drugs and the consequences on the drug formulation strategy is discussed. The integration of computational techniques in drug development is getting more and more attention with the immense progress in computational power and the pressure to improve the efficiency and reduce the overall cost of the development process. In this work the use of computer models was employed in two aspects. First, the performance of two simulations software was compared in the ability to predict the solubility of poorly soluble drugs in a fasted or fed biorelevant medium. Secondly a physiologically based pharmacokinetic (PBPK) modelling method was utilised to simulate the pharmacokinetic profiles of the same drugs. Interestingly both models produced satisfactory results for the solubility prediction of acidic drug and in the fed state however the solubility prediction of the neutral drug and in the fasted state showed some limitations. This outcome highlighted the need for improvement of in silico models for solubility prediction in particularly with the development of more sophisticated models for the integration of the multiple components present in the human intestinal fluid. Nonetheless, the successful PBPK prediction of the plasma concentration profiles confirmed the interest of this approach to be a useful tool in the drug development process. In addition the results demonstrated that for two drugs variable input of solubility had a significant influence on the in vivo plasma concentrations. Finally a PBPK approach in a special population is presented in this work. It was used to model the pharmacokinetic behaviour of a cancer molecule in patients. The objective was to reproduce the phase I dose escalation study. Due to the sparse in vitro data and the poor solubility and permeability profile of the drug, the model development was very challenging and could not be fully validated. In fact the main obstacle of the model was the suboptimal formulation and the absorption of the drug in the intestinal tract. The performance of the software to successfully predict this drug could be challenged by another PBPK software for comparison. However given the physico-chemical properties of this drug its behaviour is very likely to be drug related. This case study emphasised the difficulty of developing poorly soluble drugs.
Supervisor: Halbert, Gavin Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral