Bacterial load slopes represent biomarkers of tuberculosis therapy success, failure, and relapse.

There is an urgent need to discover biomarkers that are predictive of long-term TB treatment outcomes, since treatment is expense and prolonged to document relapse. We used mathematical modeling and machine learning to characterize a predictive biomarker for TB treatment outcomes. We computed bacterial kill rates, γf for fast- and γs for slow/non-replicating bacteria, using patient sputum data to determine treatment duration by computing time-to-extinction of all bacterial subpopulations. We then derived a γs-slope-based rule using first 8 weeks sputum data, that demonstrated a sensitivity of 92% and a specificity of 89% at predicting relapse-free cure for 2, 3, 4, and 6 months TB regimens. In comparison, current methods (two-month sputum culture conversion and the Extended-EBA) methods performed poorly, with sensitivities less than 34%. These biomarkers will accelerate evaluation of novel TB regimens, aid better clinical trial designs and will allow personalization of therapy duration in routine treatment programs.

Minocycline treatment for pulmonary Mycobacterium avium complex disease based on pharmacokinetics/pharmacodynamics and Bayesian framework mathematical models.

OBJECTIVES: Our aim was to identify the pharmacokinetic/pharmacodynamic parameters of minocycline in the hollow-fibre system (HFS) model of pulmonary Mycobacterium avium complex (MAC) and to identify the optimal clinical dose. METHODS: Minocycline MICs for 55 MAC clinical isolates from the Netherlands were determined. We also co-incubated primary isolated macrophages infected with MAC with minocycline. Next, we performed a 28 day HFS-MAC model dose-response study in which we mimicked pulmonary concentration-time profiles achieved in patients. The HFS-MAC model was sampled at intervals to determine the minocycline pharmacokinetics and MAC burden. We identified the AUC0-24/MIC ratios associated with 1.0 log10 cfu/mL kill below day 0 (stasis), defined as a bactericidal effect. We then performed 10000 Monte Carlo experiments to identify the optimal dose for a bactericidal effect in patients. RESULTS: The MIC for 50% and 90% of cumulative clinical isolates was 8 and 64 mg/L, respectively. Minocycline decreased MAC bacterial burden below stasis in primary isolated macrophages. In the HFS-MAC model, minocycline achieved a microbial kill of 3.6 log10 cfu/mL below stasis. The AUC0-24/MIC exposure associated with a bactericidal effect was 59. Monte Carlo experiments identified a minocycline susceptibility MIC breakpoint of 16 mg/L. At this proposed breakpoint, the clinical dose of 200 mg/day achieved the bactericidal effect exposure target in ∼50% of patients, while 400 mg/day achieved this in 73.6% of patients, in Monte Carlo experiments. CONCLUSIONS: Minocycline at a dose of 400 mg/day is expected to be bactericidal. We propose a clinical trial for validation.

Once-a-week tigecycline for the treatment of drug-resistant TB.

BACKGROUND: MDR-TB and XDR-TB have poor outcomes. OBJECTIVES: To examine the efficacy of tigecycline monotherapy in the hollow fibre system model of TB. METHODS: We performed pharmacokinetic/pharmacodynamic studies using tigecycline human-like concentration-time profiles in the hollow fibre system model of TB in five separate experiments using Mycobacterium tuberculosis in log-phase growth or as semi-dormant or intracellular bacilli, as monotherapy. We also compared efficacy with the isoniazid/rifampicin/pyrazinamide combination (standard therapy). We then applied extinction mathematics, morphisms and Latin hypercube sampling to identify duration of therapy with tigecycline monotherapy. RESULTS: The median tigecycline MIC for 30 M. tuberculosis clinical and laboratory isolates (67% MDR/XDR) was 2 mg/L. Tigecycline monotherapy was highly effective in killing M. tuberculosis in log-phase-growth and semi-dormant and intracellular M. tuberculosis. Once-a-week dosing had the same efficacy as daily therapy for the same cumulative dose; thus, tigecycline efficacy was linked to the AUC0-24/MIC ratio. Tigecycline replacement by daily minocycline after 4 weeks of therapy was effective in sterilizing bacilli. The AUC0-24/MIC ratio associated with optimal kill was 42.3. Tigecycline monotherapy had a maximum sterilizing effect (day 0 minus day 28) of 3.06 ±âŸ0.20 log10 cfu/mL (r2 = 0.92) compared with 3.92 ±âŸ0.45 log10 cfu/mL (r2 = 0.80) with optimized standard therapy. In our modelling, at a tigecycline monotherapy duration of 12 months, the proportion of patients with XDR-TB who reached bacterial population extinction was 64.51%. CONCLUSIONS: Tigecycline could cure patients with XDR-TB or MDR-TB who have failed recommended therapy. Once-a-week tigecycline could also replace second-line injectables in MDR-TB regimens.

A theoretical mathematical assessment of the effectiveness of coartemether in the treatment of Plasmodium falciparum malaria infection.

This study analysed the dynamics of blood stage malaria with immune response and under administration of Coartem as a combination therapy. The techniques of mathematical modelling were used in coming up and analysing the deterministic model. Sensitivity analysis and statistical approaches were used to compare model simulated treatment results with the use of Coartem and other antimalarial drugs. We sought to theoretically assess if Coartem can bring improvement in the treatment of malaria as compared to the other drugs. Our analysis and numerical results suggest that Coartem compares well with other antimalarial drug that have been on the market. However, the shortfall of our model is that it could not give good comparative results between Coartem treatment and other combination treatment schemes with similar mode of action. Our study predicted effects of different drug treatment protocols in malaria using a theoretical mathematical model, which gives an insight into potential effective treatment schemes.