Soniazid hair drug levels among drug-susceptible pulmonary tuberculosis patients: A study to monitor exposure, adherence and treatment outcomes

Abstract

Background: Tuberculosis (TB) is still a major public health problem globally and in Uganda.Therapeutic drug monitoring (TDM) for TB improves treatment outcomes and mitigates the emergence of drug resistance (DR). Despite the availability of drugs; poor adherence and attainment of suboptimal drug levels are major barriers to successful treatment outcomes, especially in sub-Saharan African countries with limited resources. Limitations to adherence may be overcome by directly observed therapy (DOT), and plasma drug levels can shed light on adherence and exposure but have limitations. Assays of antiretroviral (ARV) drugs in hair are the strongest independent predictor of virologic response in HIV-infected patients. This ARV and MDR TB hair drug level relationship could be extrapolated to first-line anti-TB drugs to monitor adherence, exposure, and treatment outcomes. Currently, there is limited data on first-line TB drug levels in hair as a tool to monitor adherence and treatment outcomes of TB patients. Main Objective: To assess Isoniazid hair drug levels among drug-susceptible pulmonary TB patients as a tool to monitor exposure, adherence, and treatment outcomes. Methods: This was a mixed design of qualitative, cross-sectional, and cohort studies of confirmed TB patients being treated with the fixed-dose combination of Isoniazid, Rifampin, Pyrazinamide, and Ethambutol (2ERHZ/4RH). The study was conducted in a framework of four sub-studies: Sub-study 1 determined the acceptability of hair harvest as a method of therapeutic drug monitoring among TB patients using a qualitative(phenomenological) design of indepth interviews, Sub-study 2 that used Liquid Chromatography-Mass Spectrometry (LCMS) to determine how hair drug levels relate to plasma drug levels at the time points of weeks: 4, 8, and 26 of adherence, Sub-study 3 that used LCMS to determine the hair drug level at which TB sputum culture conversion at the end of the intensive phase of TB treatment(8 weeks), and Sub-study 4 that used LCMS to determine how smoking, alcohol, age, and weight affect hair drug levels. Results: For Sub-study 1, there were two key findings. Key finding 1 was that hair harvest for medical testing is acceptable to the majority of TB patients provided it is harvested by a medical worker. Key finding 2 was that overall, the older age group aged at least 30 years or more is flexible when it comes to harvesting hair for medical testing using a razorblade, scissors, or modern hair-cutting machine. Many younger people; aged<30 years prefer the modern hair-cutting machine compared to razorblade or scissors. For Sub-study 2 overall, the mean ratio of hair to plasma Isoniazid hair drug levels at the week 4-time point was 1.66(SD=2.95), (95%CI: 0.55-2.76). For Sub-study 3, TB patients who archived median hair drug level of at least 0.05ng/mL (IQR: 0.1165 - 0.0314) at the end of the intensive phase of treatment (8 weeks) with first-line anti-TB drugs converted their sputum cultures to negative. For Sub-study 4, we observed no difference in Isoniazid hair drug levels per TB patient physiological and demographic factors (age, gender, and weight), smoking, and alcohol consumption on univariate and multivariate analysis. Conclusion Hair harvest for medical testing is acceptable on the condition that it is harvested by a medical worker. Tuberculosis patients aged at least 30 years are flexible on the method of hair harvest for medical testing. Patients aged below 30 years are less flexible about the preferred method of hair harvest for therapeutic drug monitoring. Isoniazid hair drug levels have a cumulative effect in hair compared to plasma. Isoniazid hair drug levels may be a predictor of TB sputum culture conversion at the end of the intensive phase. TB patient physiological factors weight, and height), smoking, and alcohol use don’t impact Isoniazid hair drug levels during TB treatment.