Full Text Available
Note: Clicking the button above will open the full text document at the original institutional repository in a new window.
Advancing tuberculosis treatment through pharmacometric modeling of drug exposure and biomarkers
Tuberculosis remains the leading cause of death from a single infectious agent globally, posing a significant threat to public health. This highlights the need for improved treatment strategies to enhance outcomes and limit the emergence of resistance. Optimizing therapy requires more reliable bioma...
Saved in:
| Main Author: | |
|---|---|
| Other Authors: | |
| Format: | Thesis |
| Language: | English |
| Published: |
Department of Medicine
2026
|
| Subjects: | |
| Tags: |
No Tags, Be the first to tag this record!
|
| _version_ | 1867613211428126720 |
|---|---|
| access_status_str | Open Access |
| author | Wijk, Marie Sjölin |
| author2 | Denti, Paolo |
| author_browse | Denti, Paolo Wijk, Marie Sjölin |
| author_facet | Denti, Paolo Wijk, Marie Sjölin |
| author_sort | Wijk, Marie Sjölin |
| collection | Thesis |
| description | Tuberculosis remains the leading cause of death from a single infectious agent globally, posing a significant threat to public health. This highlights the need for improved treatment strategies to enhance outcomes and limit the emergence of resistance. Optimizing therapy requires more reliable biomarkers and a better understanding of risk factors to enable individualized treatment approaches. In this thesis, I utilized pharmacometric modelling and data from drug-susceptible pulmonary tuberculosis patients as well as in vitro data to characterise the pharmacokinetics (PK) and pharmacodynamics (PD) of first-line anti-tuberculosis drugs and tuberculosis biomarkers. A key focus in these analyses was to investigate the impact of alcohol use, one of the major risk factors for tuberculosis infection, and drug exposure on tuberculosis treatment. Further, I investigated different ways of handling data below the lower limit of quantification (BLQ) in PK analyses. My analysis of the PK of rifampicin, isoniazid, pyrazinamide and ethambutol revealed that there is no significant effect of alcohol use on the drug exposure. This suggests that the poor tuberculosis treatment outcomes observed in individuals who consume alcohol is not driven by PK differences. Instead of intensified treatment, targeted interventions – such as enhanced support for treatment adherence – should be the focus for improving outcomes in this vulnerable population. From the same study, my analysis of time-to-positivity (TTP) data showed that individuals with higher rifampicin exposure have faster bacillary clearance. Further, I found a trend of slower bacillary clearance rate, as measured by TTP, in individuals with poor treatment outcome, suggesting a link between low rifampicin exposure in first-line tuberculosis treatment, and other variables predictive of slow bacillary clearance, to poor treatment outcome. The in vitro analysis characterized the relationship between biomarkers on solid versus liquid media. For samples with the same number of colony forming units (quantified on solid media), drug-treated and stationary phase cells had a shorter TTP (quantified on liquid media) than drug-free controls and early logarithmic phase cells, respectively. Similarly, stationary phase samples reached higher growth units and had shorter time-to-growth than early-log phase ones. This suggests the presence of a bacterial subpopulation that is differentially recovered in liquid culture. A better understanding of how TTP reflects bacterial dynamics can enhance its use as a biomarker for monitoring treatment response and optimizing the design of novel treatment regimens. Further, my investigation of different ways to handle BLQ data in PK analyses showed that, while the current gold standard method produces the least biased PK parameter estimates, it introduces numerical instability in the model. I present an alternative method in which the BLQ data are imputed, and an inflated error is applied to this data to account for the uncertainty in the imputation. This method presents a bias level similar to that of the gold standard, but far superior model stability and simpler implementation. In conclusion, by using pharmacometric modelling, I successfully identified factors affecting the PK and PD of first-line anti-tuberculosis drugs. I recommend optimized dosing to ensure adequate exposure, and further use of pharmacometric modelling to investigate the poor treatment outcomes associated with alcohol use. |
| format | Thesis |
| id | oai:open.uct.ac.za:11427/42683 |
| institution | University of Cape Town (South Africa) |
| language | eng |
| last_indexed | 2026-06-10T12:32:31.718Z |
| license_str | Not specified — see source repository |
| provenance_str_mv | Harvested via OAI-PMH from UCTD — University of Cape Town Open Access Repository |
| publishDate | 2026 |
| publishDateRange | 2026 |
| publishDateSort | 2026 |
| publisher | Department of Medicine |
| publisherStr | Department of Medicine |
| record_format | dspace |
| source_str | UCTD — University of Cape Town Open Access Repository |
| spelling | oai:open.uct.ac.za:11427/42683 Advancing tuberculosis treatment through pharmacometric modeling of drug exposure and biomarkers Wijk, Marie Sjölin Denti, Paolo tuberculosis biomarkers pharmacometric modelling Tuberculosis remains the leading cause of death from a single infectious agent globally, posing a significant threat to public health. This highlights the need for improved treatment strategies to enhance outcomes and limit the emergence of resistance. Optimizing therapy requires more reliable biomarkers and a better understanding of risk factors to enable individualized treatment approaches. In this thesis, I utilized pharmacometric modelling and data from drug-susceptible pulmonary tuberculosis patients as well as in vitro data to characterise the pharmacokinetics (PK) and pharmacodynamics (PD) of first-line anti-tuberculosis drugs and tuberculosis biomarkers. A key focus in these analyses was to investigate the impact of alcohol use, one of the major risk factors for tuberculosis infection, and drug exposure on tuberculosis treatment. Further, I investigated different ways of handling data below the lower limit of quantification (BLQ) in PK analyses. My analysis of the PK of rifampicin, isoniazid, pyrazinamide and ethambutol revealed that there is no significant effect of alcohol use on the drug exposure. This suggests that the poor tuberculosis treatment outcomes observed in individuals who consume alcohol is not driven by PK differences. Instead of intensified treatment, targeted interventions – such as enhanced support for treatment adherence – should be the focus for improving outcomes in this vulnerable population. From the same study, my analysis of time-to-positivity (TTP) data showed that individuals with higher rifampicin exposure have faster bacillary clearance. Further, I found a trend of slower bacillary clearance rate, as measured by TTP, in individuals with poor treatment outcome, suggesting a link between low rifampicin exposure in first-line tuberculosis treatment, and other variables predictive of slow bacillary clearance, to poor treatment outcome. The in vitro analysis characterized the relationship between biomarkers on solid versus liquid media. For samples with the same number of colony forming units (quantified on solid media), drug-treated and stationary phase cells had a shorter TTP (quantified on liquid media) than drug-free controls and early logarithmic phase cells, respectively. Similarly, stationary phase samples reached higher growth units and had shorter time-to-growth than early-log phase ones. This suggests the presence of a bacterial subpopulation that is differentially recovered in liquid culture. A better understanding of how TTP reflects bacterial dynamics can enhance its use as a biomarker for monitoring treatment response and optimizing the design of novel treatment regimens. Further, my investigation of different ways to handle BLQ data in PK analyses showed that, while the current gold standard method produces the least biased PK parameter estimates, it introduces numerical instability in the model. I present an alternative method in which the BLQ data are imputed, and an inflated error is applied to this data to account for the uncertainty in the imputation. This method presents a bias level similar to that of the gold standard, but far superior model stability and simpler implementation. In conclusion, by using pharmacometric modelling, I successfully identified factors affecting the PK and PD of first-line anti-tuberculosis drugs. I recommend optimized dosing to ensure adequate exposure, and further use of pharmacometric modelling to investigate the poor treatment outcomes associated with alcohol use. 2026-01-26T09:46:03Z 2026-01-26T09:46:03Z 2025 2026-01-26T09:43:15Z Thesis / Dissertation Doctoral PhD http://hdl.handle.net/11427/42683 eng application/pdf Department of Medicine Faculty of Health Sciences University of Cape Town |
| spellingShingle | tuberculosis biomarkers pharmacometric modelling Wijk, Marie Sjölin Advancing tuberculosis treatment through pharmacometric modeling of drug exposure and biomarkers |
| thesis_degree_str | Doctoral |
| title | Advancing tuberculosis treatment through pharmacometric modeling of drug exposure and biomarkers |
| title_full | Advancing tuberculosis treatment through pharmacometric modeling of drug exposure and biomarkers |
| title_fullStr | Advancing tuberculosis treatment through pharmacometric modeling of drug exposure and biomarkers |
| title_full_unstemmed | Advancing tuberculosis treatment through pharmacometric modeling of drug exposure and biomarkers |
| title_short | Advancing tuberculosis treatment through pharmacometric modeling of drug exposure and biomarkers |
| title_sort | advancing tuberculosis treatment through pharmacometric modeling of drug exposure and biomarkers |
| topic | tuberculosis biomarkers pharmacometric modelling |
| url | http://hdl.handle.net/11427/42683 |
| work_keys_str_mv | AT wijkmariesjolin advancingtuberculosistreatmentthroughpharmacometricmodelingofdrugexposureandbiomarkers |