Tuberculosis(TB) was declared to be a global emergency in 1993, with South Africa declaring it to be the country’s top health priority in 1996, but ineffective treatment strategies have led to fewer than half of all treated patients in South Africa being cured. At present,paediatric treatment remains a problem, as the antitubercular preparations of rifampicin, isoniazid and pyrazinamide, that are currently available, were not initially designed for the treatment of paediatric TB patients, providing a motivation for this project. The aim of this project is thus the development of a microemulsion dosage form for the oral delivery of RIF(Rifampicin), INH(Isoniazid) and PZA(Pyrazinamide) in combination. RIF, INH and PZA were adequately characterised with reference to the monograph standards referenced and were found to be sufficiently pure to be used in subsequent work. A chromatographic system and conditions were selected and validated as being optimal for HPLC analysis of RIF, INH and PZA in combination, with a drug partitioning method for miglyol 812 developed and validated. Ternary and pseudo-ternary phase diagrams were constructed and reported, all employing miglyol 812 as the lipid. It was undoubtedly the imwitor 308 and crillet 3 combination o/w microemulsion system that proved most successful, maintaining homogeneity on dilution. The microemulsion used in formulation comprised imwitor 308 (27.63%), crillet 3 (27.63%), miglyol 812 23.68%) and water (21.06%). The stability of RIF, INH and PZA was investigated in aqueous solution, miglyol 812, corn oil, 10%m/v cremophor RH, 5%m/v imwitor 308, 10%m/v crillet 3 and 70%m/v sorbitol solution. Trends in the stability assessments conducted on RIF, INH and PZA were noted, with slight variation depending on the formulation component being evaluated. RIF invariably demonstrated temperature and oxidation dependent degradation in all vehicles, with a definite distinction possible between samples stored at 25, 40 and 600C over a 7 day trial period. A definite advantage of storing RIF solutions under nitrogen was observed, with these solutions showing less degradation over the course of the trial, than those stored under air. INH produced a pronounced increase in the degree of degradation of RIF, whereas PZA had a negligible effect on it’s stability. INH proved to be most stable in the 70%m/v sorbitol solution with no significant oxidation or temperature dependent degradation indicated. Temperature dependent degradation was only noticable when INH was in combination with RIF, most significant in crillet 3 solution. PZA was the most stable of the three drugs, remaining relatively unaffected by temperature and the presence of air, independent of the vehicle employed, although the drug remaining did decrease slightly in the presence of RIF.Due to drug dose specifications and solubility limitations, the final formulation assessed, only contained RIF and INH, despite INH and PZA having no significant effect on the stability of each other. The solubility of PZA in the lipid and aqueous components of the microemulsion was not great enough to achieve the required 500 mg/10ml dose, while RIF and INH could achieve the respective 150mg/10ml and 100mg/10ml dose. RIF stability was improved, as anticipated, with the incorporation of RIF into the internal phase decreasing contact with INH which has been shown to affect it’s stability. RIF behaved as predicted, possessing greater stability than shown in the individual formulation components, however, INH did not, being less stable in formulation in the absence of antioxidant, than in it’s presence. A novel microemulsion formulation capable of delivering the incompatible RIF and INH in combination, with numerous microemulsion systems mapped,with the ability of being used for the delivery of other lipophilic drugs and drug combinations, was produced.The final formulation provided valuable information into possible future improvements of the microemulsion to improve drug stability.

Thesis (MScMedSc)--Stellenbosch University, 2012. / ENGLISH ABSTRACT: BACKGROUND: The emergence of mycobacterium tuberculosis resistance to first line drugs has renewed interest in second-line anti-tuberculosis drugs. Generally, Paraaminosalicylic acid (PAS) is less potent and frequently more toxic than the first line drugs. Furthermore, the pharmacokinetics of PAS in children has not been well characterized. AIMS: The aims of the present study were (1) to determine the pharmacokinetics of PAS in pediatric patients, (2) to describe the discrepancy between children and adult pharmacokinetics and the appropriate dosing regimen of PAS and (3) to investigate the potential of the second-line anti-tuberculosis drugs PAS, terizidone and ethionamide (often used as first-line drug in children) to inhibit the catalytic activities of CYP450 1A2 and 2C9. PATIENTS: Twenty two patients with drug resistant tuberculosis were included in the study. Ten patients were children with mean age of 4.2 years (range: 1 to 12 years). Twelve patients were adults with mean age of 31.3 years (range: 18 to 53). 4 children (40%) and 4 adults (33.3%) were HIV positive and were on ART. METHODS: Children received 75 mg/kg twice daily on the first visit and after two weeks they received 150 mg/kg once. Adults received a standard 4 g twice daily. Blood samples were taken at different time points after the dose. In the additional study, the inhibitory effects of PAS, ethionamide and terizidone on phenacetin O-deethylation, a marker substrate of CYP1A2 and diclofenac 4’-hydroxylation, a marker substrate of CYP2C9, were studied using human liver microsomes. RESULTS: For the 75 mg/kg dose, the mean AUC was 233.3 =g•h/ml and the mean CL was 10.4 l/h/kg. The mean of the observed Cmax of the drug was 45.4 =g/ml and the mean Tmax was 4.8 hrs. For the 150 mg/kg dose, the mean AUC of PAS was 277.9 =g•h/ml and the mean CL was 47.1 l/h/kg. The mean of the observed Cmax of the drug was 56.5 =g/ml and the mean Tmax was 4.8 hrs. On the first visit the mean AUC was 368 =g•h/ml and the mean CL was 13.2 l/h/kg. The mean of the observed Cmax of PAS was 51.3 =g/ml and the mean Tmax was 5.2 hrs. On the second visit the mean AUC was 230 =g•h/ml and the mean CL was 23.9 l/h/kg. The mean of the observed Cmax of PAS was 37.6 =g/ml and the mean Tmax was 5.2 hrs. The comparisons between pharmacokinetics profile of PAS and patients characteristics e.g. age, indicated no statistically significant differences between children (both treatment regimens) and adult patients as well as HIV positive and negative patients. In the in vitro study, all drugs demonstrated no inhibition potency towards the investigated CYP450 enzymes. CONCLUSIONS:The dose of 75 mg/kg twice daily in children appears to be appropriate to achieve serum concentration above the PAS minimum inhibitory concentration of approximately 1 =g/ml. PAS, ethionamide and terizidone are unlikely to affect the metabolism of concomitantly administered medications that are metabolized by either CYP450 1A2 and/or 2C9 isoenzymes. / AFRIKAANSE OPSOMMING: AGTERGROND: Die opkoming van eersteliniemiddel-weerstandige mycobacterium tuberculosis het opnuut belangstelling in tweedelinie-antituberkulosemiddels aangewakker. Oor die algemeen is para-aminosalisielsuur (PAS) minder kragtig en dikwels ook meer toksies. Verder is die farmakokinetika van PAS in kinders nog nie goed vasgestel nie. DOELSTELLINGS: Die doelstellings van hierdie studie was (1) om die farmakokinetika van PAS in pediatriese pasiënte vas te stel, (2) om die diskrepansie tussen kinder- en volwasse-farmakokinetika, sowel as die toepaslike doseringskedule, van PAS te beskryf en (3) om die potensiaal van die tweedeline-antituberkulosemiddels PAS, terisidoon en etioonamied (gereeld gebruik as eerste linie middels in kinders) te ondersoek wat betref hul vermoë om die katalitiese werking van CYP450 1A2 en 2C9 te inhibeer. PASIËNTE: Twee-en-twintig pasiënte met middelweerstandige tuberkulose is in hierdie studie ingesluit. Tien pasiënte was kinders met ‘n gemiddelde ouderdom van 4.2 jaar (reeks: 1 tot 12 jaar). Twaalf pasiënte was volwassenes met ‘n gemiddelde ouderdom van 31.3 jaar (reeks: 18 tot 53 jaar). 4 kinders (40%) en 4 volwassenes (33.3%) was MIV positief en was op TRM’s. METODES: Kinders het 75 mg/kg twee maal daaliks gedurende die eerste besoek ontvang en 150 mg/kg een maal ná twee weke ontvang. Volwassenes het ‘n standaarddosis van 4 g twee maal daagliks ontvang. Bloedmonsters is op verskillende tye ná die dosering geneem. In die addisionele studie is in die inhiberende effekte van PAS, etioonamied en terisidoon op fenasetien-O-deëtilering, ‘n merkersubstraat van CYP1A2 en diklofenak-4’-hidroksilasie, ‘n merkersubstraat van CYP2C9, ondersoek deur gebruik te maak van menslike lewermikrosome. RESULTATE: Vir die 75 mg/kg dosis was die gemiddelde area-onder-die-kurwe (AOK) 233.3 =g•h/ml en die gemiddelde middelopruiming (CL) 10.4 l/h/kg. Die gemiddelde geobserveerde Cmaks van die middel was 45.4 =g/ml en die gemiddelde Tmaks was 4.8 h. Vir die 150 mg/kg dosering was die gemiddelde AOK van PAS 277.9 =g•h/ml en die gemiddelde CL 47.1 l/h/kg. Die gemiddelde geobserveerde Cmaks van die middel was 56.5 =g/ml en die gemiddelde Tmaks was 4.8 h. Gedurende die eerste besoek was die AOK 368 =g•h/ml en die gemiddelde CL was 13.2 l/h/kg. Die gemiddelde geobserveerde Cmaks van PAS was 51.3 =g/ml en die gemiddelde Tmaks was 5.2 h. Gedurende die tweede besoek was die gemiddelde AOK 230 =g•h/ml en die gemiddelde CL 23.9 l/h/kg. Die gemiddelde geobserveerde Cmaks van PAS was 37.6 =g/ml en die gemiddelde Tmaks was 5.2 h. Die vergelyking van PAS-farmakokinetika en eienskappe van die pasiënte het geen statisties beduidende verskille in die gemiddelde AOK tussen kinders (op albei doserings) en volwassenes getoon nie. Met die in vitrostudie het geen van die middels inhibisie-werking teenoor die CYP450-ensieme wat ondersoek is, getoon nie. GEVOLGTREKKINGS: Die gevolgtrekking kan gemaak word dat die dosering van 75 mg/kg twee maal daagliks voldoende is om serumkonsentrasies wat bo PAS se minimum inhiberende konsentrasie van 1 =g/ml te bereik. Dit is onwaarskynlik dat PAS, etioonamied en terisidoon die metabolisme van gelyktydig-toegediende medikasies, wat op hul beurt deur die CYP240-isoënsieme 1A2 en/of 2C9 gemetaboliseer word, sal affekteer. / Division of Pharmacology, Stellenbosch University / National Research Foundation (NRF) grant generously offered by Professor Donald Grant