Waters, Robert K.; (2001) The development of high resolution techniques for the surveillance of medicines. Doctoral thesis (Ph.D.), University College London.

Text The_development_of_high_resolu.pdf Download (10MB)

Abstract

There is a continuing requirement for the development of rapid and accurate methods for the identification and quantification of drugs particularly in generic pharmaceutical formulations and forensic samples. The use of surveillance methods, (post-market), is of interest to the pharmaceutical industry and regulatory authorities alike. The rise of generic and parallel imported drug products means there are many more sources of the ever increasing range of drug formulations. In the following work a non-buffered gradient UPLC system was considered as the basis of a rapid drug screening method. The system was originally developed by Gill et al for use in The Forensic Science Service, UK. Here the system has been developed further and tested as a surveillance technique on 'real' pharmaceutical samples. Gradient and isocratic versions of the system were set up on a modular Gilson HPLC (column: Inertsil ODS-2), (System 1) and on a Waters Millennium system (column: Waters Symmetry), (System 2). A database of gradient retention indices for 178 and 237 was obtained on Systems 1 and 2 respectively. System predictability and inter-laboratory transfer of data were investigated. A group of 28 barbiturates were run on System I. The system demonstrated good repeatability with retention index values showing coefficients of variation of less than 0.46% for gradient and 0.23% for isocratic runs. Retention values were correlated with a range of structural and physical parameters to test the selective behaviour of the system. Correlation with molecular connectivity values was poor as was the correlation with GLC retention index values (r = 0.45). Correlation of the HPLC retention indices with molar volume (r = 0.831) and Log P values (r = 0.891) suggested these were factors affecting retention. Excellent correlation between the experimental retention indices and calculated indices, derived by the addition of substituent component values to the barbituric acid core, was obtained. This demonstrated the internal consistency and predictability of the system. Using System 2, variations in eluent pH, column temperature and eluent composition were used to demonstrate the excellent robustness of the system. The repeatability of the 16 nitro-n-alkanes used to assign index values to the retention times of four acidic and four basic drugs, were not affected by the applied conditions. The mean standard deviation for these retention times was less than 0.1 minute. The variation in retention indices for the acidic and basic drugs was monitored with changing conditions. Gradient retention indices of 73 drugs run on System 1 common to those run on System 2 gave a correlation of r = 0.986. A data set of 237 drugs run on System 2 was used to generate a database of retention indices to demonstrate inter-laboratory transfer. These retention indices were obtained from the equations that described the correlations between System 2 retention indices and retention data given by nine other groups on various systems. System 2 utilised a photodiode array detector so that ultraviolet spectra, as well as retention index data could be used to identify a drug. A modern polymeric column that could tolerate high pH was investigated as an orthoganol technique. It was demonstrated that for some drugs a change in ionisation state led to a different retention index and UV spectrum. System 2 was used to test the surveillance procedure on sample supplied by a pharmaceutical exporter. Four powders for injection were considered. The active ingredient in each case was identified and quantified, the 95% confidence limits for the quantification fell within the BP limits for each drug. The decomposition of a new formulation of 5-aminolevulinic acid was monitored using the system demonstrating its flexibility for other analytical tasks. The use of Near Infrared techniques for rapid, non-destructive analysis of drug samples was considered. In particular the role of glass in the transflectance of light was investigated by reflecting incident light on glass vials containing mercury and then correcting the spectra of the drug in glass with the result.

Download activity - last month

Export as XMLCSVJSON

Download activity - last 12 months

Export as JSONXMLCSV

Downloads by country - last 12 months

Export as JSONCSVXML

Archive Staff Only

View Item